Expert system: Difference between revisions

Line 1:

{{Short description|Computer system emulating ~~the decision-making ability of a~~ human expert}}

{{For|the science fiction book series by Adrian Tchaikovsky|Expert Systems (series)}}

[[File:Symbolics3640 Modified.JPG|200px|thumbnail|right|A [[Symbolics]] 3640 [[Lisp machine]]: an early (1984) platform for expert systems]]

In [[artificial intelligence]] (AI), an '''expert system''' is a computer system emulating the decision-making ability of a human [[expert]].<ref name="Jackson1998">{{cite book |last1=Jackson |first1=Peter |year=1998 |title=Introduction To Expert Systems |publisher=Addison Wesley |edition=3 |isbn=978-0-201-87686-4 |page=2}}</ref>

Expert systems are designed to solve complex problems by [[Automated reasoning system|reasoning]] through bodies of knowledge, represented mainly as [[Rule-based system|if–then rules]] rather than through conventional [[procedural programming]] code.<ref>{{cite web |url=https://www.pcmag.com/encyclopedia_term/0,2542,t=conventional+programming&i=40325,00.asp |title=Conventional programming |publisher=Pcmag.com |access-date=2013-09-15 |archive-date=2012-10-14 |archive-url=https://web.archive.org/web/20121014124656/http://www.pcmag.com/encyclopedia_term/0%2C2542%2Ct%3Dconventional+programming%26i%3D40325%2C00.asp |url-status=dead}}</ref> Expert systems were among the first truly successful forms of AI software.<ref name="Simon & Schuster">{{cite book |last1=Russell |first1=Stuart |last2=Norvig |first2=Peter |title=Artificial Intelligence: A Modern Approach|date=1995 |publisher=Simon & Schuster |isbn=978-0-13-103805-9 |pages=22–23|url=http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf|access-date=14 June 2014 |archive-url=https://web.archive.org/web/20140505045226/http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf |archive-date=5 May 2014|url-status=dead}}</ref>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}} They were created in the 1970s and then proliferated in the 1980s,<ref name="durkinhistory">{{cite book |first= Cornelius T. |last= Leondes |title=Expert systems: the technology of knowledge management and decision making for the 21st century |year=2002 |isbn=978-0-12-443880-4 |pages=1–22}}</ref> being then widely regarded as the future of AI — before the advent of successful [[artificial neural network]]s.<ref>{{Cite news |date=2024-07-16 |title=A short history of AI |url=https://www.economist.com/schools-brief/2024/07/16/a-short-history-of-ai|access-date=2024-08-14 |newspaper=[[The Economist]]|language=en}}</ref>

Expert systems are designed to solve complex problems by [[Automated reasoning system|reasoning]] through bodies of knowledge, represented mainly as [[Rule-based system|if–then rules]] rather than through conventional [[procedural programming]] code.<ref>{{cite web |url=https://www.pcmag.com/encyclopedia_term/0,2542,t=conventional+programming&i=40325,00.asp |title=Conventional programming |publisher=Pcmag.com |access-date=2013-09-15 |archive-date=2012-10-14 |archive-url=https://web.archive.org/web/20121014124656/http://www.pcmag.com/encyclopedia_term/0%2C2542%2Ct%3Dconventional+programming%26i%3D40325%2C00.asp |url-status=dead}}</ref> Expert systems were among the first truly successful forms of AI software.<ref name="Simon & Schuster">{{cite book |last1=Russell |first1=Stuart |last2=Norvig |first2=Peter |title=Artificial Intelligence: A Modern Approach|date=1995 |publisher=Simon & Schuster |isbn=978-0-13-103805-9 |pages=22–23|url=http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf|access-date=14 June 2014 |archive-url=https://web.archive.org/web/20140505045226/http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf |archive-date=5 May 2014|url-status=dead}}</ref>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}} They were created in the 1970s and then proliferated in the 1980s,<ref name="durkinhistory">{{cite book |first= Cornelius T. |last= Leondes |title=Expert systems: the technology of knowledge management and decision making for the 21st century |year=2002 |isbn=978-0-12-443880-4 |pages=1–22}}</ref> being then widely regarded as the future of AI — before the advent of successful [[artificial neural network]]s.<ref>{{Cite news |date=2024-07-16 |title=A short history of AI |url=https://www.economist.com/schools-brief/2024/07/16/a-short-history-of-ai |access-date=2024-08-14 |newspaper=[[The Economist]] |language=en |archive-date=2024-08-13 |archive-url=https://web.archive.org/web/20240813210432/https://www.economist.com/schools-brief/2024/07/16/a-short-history-of-ai |url-status=live }}</ref>

An expert system is divided into two subsystems: 1) a ''[[knowledge base]]'', which represents facts and rules; and 2) an ''[[inference engine]]'', which applies the rules to the known facts to deduce new facts, and can include explaining and debugging abilities.

== History ==

=== Early development ===

Soon after the dawn of modern computers in the late 1940s and early 1950s, researchers started realizing the immense potential these machines had for modern society. One of the first challenges was to make such machines able to “think” like humans – in particular, making these machines able to make important decisions the way humans do. The ~~medical–healthcare~~ field presented the tantalizing challenge of enabling these machines to make medical diagnostic decisions.<ref name="CADsurvey">{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=A Systematic Survey of Computer-Aided Diagnosis in Medicine: Past and Present Developments |journal=Expert Systems with Applications |volume=138 |article-number=112821 |date=2019 |doi=10.1016/j.eswa.2019.112821 |s2cid=199019309}}</ref>

Soon after the dawn of modern computers in the late 1940s and early 1950s, researchers started realizing the immense potential these machines had for modern society. One of the first challenges was to make such machines able to “think” like humans – in particular, making these machines able to make important decisions the way humans do. The medical–[[Health care|healthcare]] field presented the tantalizing challenge of enabling these machines to make medical diagnostic decisions.<ref name="CADsurvey">{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=A Systematic Survey of Computer-Aided Diagnosis in Medicine: Past and Present Developments |journal=Expert Systems with Applications |volume=138 |article-number=112821 |date=2019 |doi=10.1016/j.eswa.2019.112821 |s2cid=199019309}}</ref>

Thus, in the late 1950s, right after the information age had fully arrived, researchers started experimenting with the prospect of using computer technology to emulate human decision making. For example, biomedical researchers started creating computer-aided systems for diagnostic applications in medicine and biology. These early diagnostic systems used patients’ symptoms and laboratory test results as inputs to generate a diagnostic outcome.<ref>{{cite journal |vauthors=Ledley RS, and Lusted LB |title=Reasoning foundations of medical diagnosis |journal=Science |date=1959 |volume=130 |issue=3366 |pages=9–21 |doi=10.1126/science.130.3366.9 |pmid=13668531 |bibcode=1959Sci...130....9L}}</ref><ref>{{cite journal |vauthors=Weiss SM, Kulikowski CA, Amarel S, Safir A |title=A model-based method for computer-aided medical decision-making |journal=Artificial Intelligence |date=1978 |volume=11 |issue=1–2 |pages=145–172 |doi=10.1016/0004-3702(78)90015-2 |citeseerx=10.1.1.464.3183}}</ref>

These systems were often described as the early forms of expert systems. However, researchers realized that there were significant limits when using traditional methods such as flow charts,<ref>{{cite journal |vauthors=Schwartz WB |date=1970 |title=Medicine and the computer: the promise and problems of change |journal=New England Journal of Medicine |volume=283 |issue=23 |pages=1257–1264 |doi=10.1056/NEJM197012032832305 |pmid=4920342}}</ref>

<ref>{{cite journal |vauthors=Bleich HL |date=1972 |title=Computer-based consultation: Electrolyte and acid-base disorders |journal=The American Journal of Medicine |volume=53 |issue=3 |pages=285–291 |doi=10.1016/0002-9343(72)90170-2 |pmid=4559984}}</ref> statistical pattern matching,<ref>{{cite journal |vauthors=Rosati RA, McNeer JF, Starmer CF, Mittler BS, Morris JJ, and Wallace AG |date=1975 |title=A new information system for medical practice |journal=Archives of Internal Medicine |volume=135 |issue=8 |pages=1017–1024 |doi=10.1001/archinte.1975.00330080019003 |pmid=1156062}}</ref> or probability theory.<ref>{{cite journal |vauthors=Gorry GA, Kassirer JP, Essig A, and Schwartz WB |date=1973 |title=Decision analysis as the basis for computer-aided management of acute renal failure |journal=The American Journal of Medicine |volume=55 |issue=4 |pages=473–484 |doi=10.1016/0002-9343(73)90204-0 |pmid=4582702 |s2cid=17448496}}</ref><ref>{{cite journal |vauthors=Szolovits P, Patil RS, and Schwartz WB |date=1988 |title=Artificial intelligence in medical diagnosis |journal=Annals of Internal Medicine |volume=108 |issue=1 |pages=80–87 |doi=10.7326/0003-4819-108-1-80 |pmid=3276267 |s2cid=46410202}}</ref>

<ref>{{cite journal |vauthors=Bleich HL |date=1972 |title=Computer-based consultation: Electrolyte and acid-base disorders |journal=The American Journal of Medicine |volume=53 |issue=3 |pages=285–291 |doi=10.1016/0002-9343(72)90170-2 |pmid=4559984}}</ref> statistical [[pattern matching]],<ref>{{cite journal |vauthors=Rosati RA, McNeer JF, Starmer CF, Mittler BS, Morris JJ, and Wallace AG |date=1975 |title=A new information system for medical practice |journal=Archives of Internal Medicine |volume=135 |issue=8 |pages=1017–1024 |doi=10.1001/archinte.1975.00330080019003 |pmid=1156062}}</ref> or probability theory.<ref>{{cite journal |vauthors=Gorry GA, Kassirer JP, Essig A, and Schwartz WB |date=1973 |title=Decision analysis as the basis for computer-aided management of acute renal failure |journal=The American Journal of Medicine |volume=55 |issue=4 |pages=473–484 |doi=10.1016/0002-9343(73)90204-0 |pmid=4582702 |s2cid=17448496}}</ref><ref>{{cite journal |vauthors=Szolovits P, Patil RS, and Schwartz WB |date=1988 |title=Artificial intelligence in medical diagnosis |journal=Annals of Internal Medicine |volume=108 |issue=1 |pages=80–87 |doi=10.7326/0003-4819-108-1-80 |pmid=3276267 |s2cid=46410202}}</ref>

=== Formal introduction and later developments ===

Line 18:

Line 20:

This previous situation gradually led to the development of expert systems, which used knowledge-based approaches. These expert systems in medicine were the [[MYCIN]] expert system,<ref>{{cite journal |vauthors=Shortliffe EH, and Buchanan BG |title=A model of inexact reasoning in medicine |journal=Mathematical Biosciences |date=1975 |volume=23 |issue=3–4 |pages=351–379 |doi=10.1016/0025-5564(75)90047-4 |s2cid=118063112}}</ref> the [[Internist-I]] expert system<ref>{{cite journal |vauthors=Miller RA, Pople Jr HE, and Myers JD |date=1982 |title=Internist-I, an experimental computer-based diagnostic consultant for general internal medicine |journal=New England Journal of Medicine |volume=307 |issue=8 |pages=468–476 |doi=10.1056/NEJM198208193070803 |pmid=7048091}}</ref> and later, in the middle of the 1980s, the [[CADUCEUS (expert system)|CADUCEUS]].<ref>{{cite book |last1=Feigenbaum |first1=Edward |pages=1–275 |title=The fifth generation |year=1984 |publisher=Addison-Wesley |isbn=978-0451152640 |first2=Pamela |last2=McCorduck}}</ref>

Expert systems were formally introduced around 1965 by the [[Stanford University|Stanford]] Heuristic Programming Project led by [[Edward Feigenbaum]], who is sometimes termed the "father of expert systems";<ref>{{Cite web |last=Joseph |first=Staney |date=2023-10-30 |title=The Diversity of Artificial Intelligence: How Edward Feigenbaum Developed the Expert Systems |url=https://medium.com/@staneyjoseph.in/the-diversity-of-artificial-intelligence-how-edward-feigenbaum-developed-the-expert-systems-8fe57350efb8 |access-date=2024-01-26 |website=Medium |language=en}}</ref> other key early contributors were Bruce Buchanan and Randall Davis. The Stanford researchers tried to identify domains where expertise was highly valued and complex, such as diagnosing infectious diseases ([[Mycin]]) and identifying unknown organic molecules ([[Dendral]]).<ref>{{cite book |last1=Lea |first1=Andrew S. |year=2023 |title=Digitizing Diagnosis: Medicine, Minds, and Machines in Twentieth-Century America |publisher=Johns Hopkins University Press |pages=1–256 |isbn=978-1421446813}}</ref> The idea that "intelligent systems derive their power from the knowledge they possess rather than from the specific formalisms and inference schemes they use"<ref>Edward Feigenbaum, 1977. Paraphrased by Hayes-Roth, et al.</ref> – as Feigenbaum said – was at the time a significant step forward, since the past research had been focused on heuristic computational methods, culminating in attempts to develop very general-purpose problem solvers (foremostly the conjunct work of [[Allen Newell]] and [[Herbert A. Simon|Herbert Simon]]).<ref>{{cite book |last1=Hayes-Roth |first1=Frederick |last2=Waterman |first2=Donald |last3=Lenat |first3=Douglas |author3-link=Douglas Lenat |year=1983 |pages=[https://archive.org/details/buildingexpertsy00temd/page/6 6–7] |title=Building Expert Systems |publisher=Addison-Wesley |isbn=978-0-201-10686-2 |url=https://archive.org/details/buildingexpertsy00temd/page/6}}</ref> Expert systems became some of the first truly successful forms of [[artificial intelligence]] (AI) software.<ref name="Simon & Schuster"/>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}}

Expert systems were formally introduced around 1965 by the [[Stanford University|Stanford]] Heuristic Programming Project led by [[Edward Feigenbaum]], who is sometimes termed the "father of expert systems";<ref>{{Cite web |last=Joseph |first=Staney |date=2023-10-30 |title=The Diversity of Artificial Intelligence: How Edward Feigenbaum Developed the Expert Systems |url=https://medium.com/@staneyjoseph.in/the-diversity-of-artificial-intelligence-how-edward-feigenbaum-developed-the-expert-systems-8fe57350efb8 |access-date=2024-01-26 |website=Medium |language=en |archive-date=2024-01-26 |archive-url=https://web.archive.org/web/20240126214030/https://medium.com/@staneyjoseph.in/the-diversity-of-artificial-intelligence-how-edward-feigenbaum-developed-the-expert-systems-8fe57350efb8 |url-status=live }}</ref> other key early contributors were Bruce Buchanan and Randall Davis. The Stanford researchers tried to identify domains where expertise was highly valued and complex, such as diagnosing infectious diseases ([[Mycin]]) and identifying unknown organic molecules ([[Dendral]]).<ref>{{cite book |last1=Lea |first1=Andrew S. |year=2023 |title=Digitizing Diagnosis: Medicine, Minds, and Machines in Twentieth-Century America |publisher=Johns Hopkins University Press |pages=1–256 |isbn=978-1421446813}}</ref> The idea that "intelligent systems derive their power from the knowledge they possess rather than from the specific formalisms and inference schemes they use"<ref>Edward Feigenbaum, 1977. Paraphrased by Hayes-Roth, et al.</ref> – as Feigenbaum said – was at the time a significant step forward, since the past research had been focused on heuristic computational methods, culminating in attempts to develop very general-purpose problem solvers (foremostly the conjunct work of [[Allen Newell]] and [[Herbert A. Simon|Herbert Simon]]).<ref>{{cite book |last1=Hayes-Roth |first1=Frederick |last2=Waterman |first2=Donald |last3=Lenat |first3=Douglas |author3-link=Douglas Lenat |year=1983 |pages=[https://archive.org/details/buildingexpertsy00temd/page/6 6–7] |title=Building Expert Systems |publisher=Addison-Wesley |isbn=978-0-201-10686-2 |url=https://archive.org/details/buildingexpertsy00temd/page/6}}</ref> Expert systems became some of the first truly successful forms of [[artificial intelligence]] (AI) software.<ref name="Simon & Schuster"/>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}}

Research on expert systems was also active in Europe. In the US, the focus tended to be on the use of [[Production system (computer science)|production rule systems]], first on systems hard coded on top of [[Lisp (programming language)|Lisp]] programming environments and then on expert system shells developed by vendors such as [[IntelliCorp (software)|Intellicorp]]. In Europe, research focused more on systems and expert systems shells developed in [[Prolog]]. The advantage of Prolog systems was that they employed a form of [[rule-based system|rule-based programming]] that was based on [[logic programming|formal logic]].<ref>George F. Luger and William A. Stubblefield, Benjamin/Cummings Publishers, Rule-Based Expert System Shell: example of code using the Prolog rule-based expert system shell</ref><ref>[http://promethee.philo.ulg.ac.be/engdep1/download/prolog/htm_docs/prolog.htm A. Michiels] {{Webarchive|url=https://web.archive.org/web/20120402132354/http://promethee.philo.ulg.ac.be/engdep1/download/prolog/htm_docs/prolog.htm |date=2012-04-02}}, Université de Liège, Belgique: "PROLOG, the first declarative language</ref>

One such early expert system shell based on Prolog was APES.<ref name="APES">{{citation |url=https://www.ojp.gov/ncjrs/virtual-library/abstracts/investigating-apes-augmented-prolog-expert-system |title=Investigating with APES (Augmented Prolog Expert System) |access-date=2024-01-03}}</ref>

One such early expert system shell based on [[Prolog]] was APES.<ref name="APES">{{citation |url=https://www.ojp.gov/ncjrs/virtual-library/abstracts/investigating-apes-augmented-prolog-expert-system |title=Investigating with APES (Augmented Prolog Expert System) |access-date=2024-01-03 |archive-date=2024-01-03 |archive-url=https://web.archive.org/web/20240103115654/https://www.ojp.gov/ncjrs/virtual-library/abstracts/investigating-apes-augmented-prolog-expert-system |url-status=live }}</ref>

One of the first use cases of [[Prolog]] and APES was in the legal area namely, the encoding of a large portion of the British Nationality Act. Lance Elliot wrote: "The British Nationality Act was passed in 1981 and shortly thereafter was used as a means of showcasing the efficacy of using Artificial Intelligence (AI) techniques and technologies, doing so to explore how the at-the-time newly enacted statutory law might be encoded into a computerized logic-based formalization. A now oft-cited research paper entitled “The British Nationality Act as a Logic Program” was published in 1986 and subsequently became a hallmark for subsequent work in AI and the law."<ref name="AI & Law">

One of the first use cases of Prolog and APES was in the legal area namely, the encoding of a large portion of the British Nationality Act. Lance Elliot wrote: "[[British Nationality Act 1981|The British Nationality Act]] was passed in 1981 and shortly thereafter was used as a means of showcasing the efficacy of using Artificial Intelligence (AI) techniques and technologies, doing so to explore how the at-the-time newly enacted statutory law might be encoded into a computerized logic-based formalization. A now oft-cited research paper entitled “The British Nationality Act as a Logic Program” was published in 1986 and subsequently became a hallmark for subsequent work in AI and the law."<ref name="AI & Law">

{{citation |url=https://lance-eliot.medium.com/ai-law-british-nationality-act-unexpectedly-spurred-ai-and-law-404aea03386a |title=AI & Law: British Nationality Act Unexpectedly Spurred AI And Law |date=17 April 2021 |access-date=2023-11-13}}</ref><ref name="BNA">{{cite journal |doi=10.1145/5689.5920 |author=M.J. Sergot and F. Sadri and R.A. Kowalski and F. Kriwaczek and P. Hammond and H.T. Cory |date=May 1986 |title=The British Nationality Act as a Logic Program |journal=Communications of the ACM |volume=29 |number=5 |pages=370–386}}</ref>

{{citation |last1=Eliot |first1=Lance |url=https://lance-eliot.medium.com/ai-law-british-nationality-act-unexpectedly-spurred-ai-and-law-404aea03386a |title=AI & Law: British Nationality Act Unexpectedly Spurred AI And Law |work=Medium |date=17 April 2021 |access-date=2023-11-13}}</ref><ref name="BNA">{{cite journal |doi=10.1145/5689.5920 |author=M.J. Sergot and F. Sadri and R.A. Kowalski and F. Kriwaczek and P. Hammond and H.T. Cory |date=May 1986 |title=The British Nationality Act as a Logic Program |journal=Communications of the ACM |volume=29 |number=5 |pages=370–386}}</ref>

In the 1980s, expert systems proliferated. Universities offered expert system courses and two-thirds of the [[Fortune 500]] companies applied the technology in daily business activities.<ref name=durkinhistory/><ref>Durkin, J. Expert Systems: Catalog of Applications. Intelligent Computer Systems, Inc., Akron, OH, 1993.</ref> Interest was international with the [[Fifth generation computer|Fifth Generation Computer Systems project]] in Japan and increased research funding in Europe.

In 1981, the first [[IBM PC]], with the [[IBM PC DOS|PC DOS]] operating system, was introduced.<ref>{{Cite web |title=The IBM PC - CHM Revolution |url=https://www.computerhistory.org/revolution/personal-computers/17/301 |access-date=2024-01-26 |website=www.computerhistory.org}}</ref> The imbalance between the low cost of the relatively powerful chips in the PC, compared to the much more expensive cost of processing power in the mainframes that dominated the corporate IT world at the time, created a new type of architecture for corporate computing, termed the [[client–server model]].<ref>{{cite book |last1=Orfali |first1=Robert |year=1996 |title=The Essential Client/Server Survival Guide |publisher=Wiley Computer Publishing |location=New York |isbn=978-0-471-15325-2 |pages=[https://archive.org/details/essentialclients00orfa/page/1 1–10] |url=https://archive.org/details/essentialclients00orfa/page/1}}</ref> Calculations and reasoning could be performed at a fraction of the price of a mainframe using a PC. This model also enabled business units to bypass corporate IT departments and directly build their own applications. As a result, client-server had a tremendous impact on the expert systems market. Expert systems were already outliers in much of the business world, requiring new skills that many IT departments did not have and were not eager to develop. They were a natural fit for new PC-based shells that promised to put application development into the hands of end users and experts. Until then, the main development environment for expert systems had been high end [[Lisp machine]]s from [[Xerox]], [[Symbolics]], and [[Texas Instruments]]. With the rise of the PC and client-server computing, vendors such as Intellicorp and Inference Corporation shifted their priorities to developing PC-based tools. Also, new vendors, often financed by [[venture capital]] ~~(such as Aion Corporation, [[Neuron Data]], Exsys, [[VP-Expert]], and many others~~<ref>{{cite book |last=Hurwitz |first=Judith |title=Smart or Lucky: How Technology Leaders Turn Chance into Success |year=2011 |publisher=John Wiley & Son |isbn=978-1118033784 |page=164|url=https://books.google.com/books?id=3KrTQzQHl7AC&q=expert+systems+failed+to+live+up+to+hype&pg=PA164|access-date=29 November 2013}}</ref><ref>{{cite journal |first=Robert J. |last=Dunn |date=September 30, 1985 |title=Expandable Expertise for Everyday Users |page=30 |journal=InfoWorld |volume=7 |issue=39 |url=https://books.google.com/books?id=iS8EAAAAMBAJ&pg=PA30 |access-date=2011-03-13}}</ref>), started appearing regularly.

In 1981, the first [[IBM PC]], with the [[IBM PC DOS|PC DOS]] operating system, was introduced.<ref>{{Cite web |title=The IBM PC - CHM Revolution |url=https://www.computerhistory.org/revolution/personal-computers/17/301 |access-date=2024-01-26 |website=www.computerhistory.org |archive-date=2024-01-26 |archive-url=https://web.archive.org/web/20240126214029/https://www.computerhistory.org/revolution/personal-computers/17/301 |url-status=live }}</ref> The imbalance between the low cost of the relatively powerful chips in the PC, compared to the much more expensive cost of processing power in the mainframes that dominated the corporate IT world at the time, created a new type of architecture for corporate computing, termed the [[client–server model]].<ref>{{cite book |last1=Orfali |first1=Robert |year=1996 |title=The Essential Client/Server Survival Guide |publisher=Wiley Computer Publishing |location=New York |isbn=978-0-471-15325-2 |pages=[https://archive.org/details/essentialclients00orfa/page/1 1–10] |url=https://archive.org/details/essentialclients00orfa/page/1}}</ref> Calculations and reasoning could be performed at a fraction of the price of a mainframe using a PC. This model also enabled business units to bypass corporate IT departments and directly build their own applications. As a result, client-server had a tremendous impact on the expert systems market. Expert systems were already outliers in much of the business world, requiring new skills that many IT departments did not have and were not eager to develop. They were a natural fit for new PC-based shells that promised to put application development into the hands of end users and experts. Until then, the main development environment for expert systems had been high end [[Lisp machine]]s from [[Xerox]], [[Symbolics]], and [[Texas Instruments]]. With the rise of the PC and client-server computing, vendors such as Intellicorp and Inference Corporation shifted their priorities to developing PC-based tools. Also, new vendors, often financed by [[venture capital]]<ref>{{cite book |last=Hurwitz |first=Judith |title=Smart or Lucky: How Technology Leaders Turn Chance into Success |year=2011 |publisher=John Wiley & Son |isbn=978-1118033784 |page=164|url=https://books.google.com/books?id=3KrTQzQHl7AC&q=expert+systems+failed+to+live+up+to+hype&pg=PA164|access-date=29 November 2013}}</ref><ref>{{cite journal |first=Robert J. |last=Dunn |date=September 30, 1985 |title=Expandable Expertise for Everyday Users |page=30 |journal=InfoWorld |volume=7 |issue=39 |url=https://books.google.com/books?id=iS8EAAAAMBAJ&pg=PA30 |access-date=2011-03-13}}</ref> started appearing regularly.

The first expert system to be used in a design capacity for a large-scale product was the Synthesis of Integral Design (SID) software program, developed in 1982. Written in [[Lisp (programming language)|Lisp]], SID generated 93% of the [[VAX 9000]] CPU logic gates.<ref name="SIDDTJ">{{cite journal |last1=Gibson |first1=Carl S. |display-authors=etal |title=VAX 9000 Series |journal=Digital Technical Journal of Digital Equipment Corporation |volume=2 |issue=4, Fall 1990 |pages=118–129}}</ref> Input to the software was a set of rules created by several expert logic designers. SID expanded the rules and generated software [[logic synthesis]] routines many times the size of the rules themselves. Surprisingly, the combination of these rules resulted in an overall design that exceeded the capabilities of the experts themselves, and in many cases out-performed the human counterparts. While some rules contradicted others, top-level control parameters for speed and area provided the tie-breaker. The program was highly controversial but used nevertheless due to project budget constraints. It was terminated by logic designers after the VAX 9000 project completion.

Line 34:

Line 36:

During the years before the middle of the 1970s, the expectations of what expert systems can accomplish in many fields tended to be extremely optimistic. At the start of these early studies, researchers were hoping to develop entirely automatic (i.e., completely computerized) expert systems. The expectations of people of what computers can do were frequently too idealistic. This situation radically changed after [[Richard M. Karp]] published his breakthrough paper: “Reducibility among Combinatorial Problems” in the early 1970s.<ref>{{cite book |last1=Karp |first1=Richard M. |year=1972 |chapter=Reducibility Among Combinatorial Problems |chapter-url=http://www.cs.berkeley.edu/~luca/cs172/karp.pdf |title=Complexity of Computer Computations |editor1-last=Miller |editor1-first=R. E. |editor2-last=Thatcher |editor2-first=J. W. |publisher=New York: Plenum |pages=85–103 |access-date=2020-01-24 |archive-date=2011-06-29 |url-status=dead |archive-url=https://web.archive.org/web/20110629023717/http://www.cs.berkeley.edu/~luca/cs172/karp.pdf}}</ref> Thanks to Karp's work, together with other scholars, like Hubert L. Dreyfus,<ref>{{cite book |author=Hubert L. Dreyfus |title=What Computers Still Can't Do |publisher=Cambridge, Massachusetts: The MIT Press |year=1972}}</ref> it became clear that there are certain limits and possibilities when one designs computer algorithms. His findings describe what computers can do and what they cannot do. Many of the computational problems related to this type of expert systems have certain pragmatic limits. These findings laid down the groundwork that led to the next developments in the field.<ref name="CADsurvey"/>

In the 1990s and beyond, the term ''expert system'' and the idea of a standalone AI system mostly dropped from the IT lexicon. There are two interpretations of this. One is that "expert systems failed": the IT world moved on because expert systems did not deliver on their over hyped promise.<ref>{{Cite web |url=http://www.ainewsletter.com/newsletters/aix_0501.htm#w |title=AI Expert Newsletter: W is for Winter |access-date=2013-11-29 |archive-url=https://web.archive.org/web/20131109201636/http://www.ainewsletter.com/newsletters/aix_0501.htm#w |archive-date=2013-11-09 |url-status=dead}}</ref><ref>{{cite journal |last1=Leith |first1=P. |date=2010 |url=http://ejlt.org//article/view/14/1 |title=The rise and fall of the legal expert system |journal=European Journal of Law and Technology |volume=1 |issue=1 |access-date=2020-01-24 |url-status=dead |archive-date=2016-03-04 |archive-url=https://web.archive.org/web/20160304124042/http://ejlt.org//article/view/14/1}}</ref> The other is the mirror opposite, that expert systems were simply victims of their success: as IT professionals grasped concepts such as rule engines, such tools migrated from being standalone tools for developing special purpose ''expert'' systems, to being one of many standard tools.<ref>{{cite journal |last1=Haskin |first1=David |date=January 16, 2003 |title=Years After Hype, 'Expert Systems' Paying Off for Some |journal=Datamation |url=http://www.datamation.com/netsys/article.php/1570851/Years-After-Hype-Expert-Systems-Paying-Off-For-Some.htm|access-date=29 November 2013}}</ref> Other researchers suggest that Expert Systems caused inter-company power struggles when the IT organization lost its exclusivity in software modifications to users or Knowledge Engineers.<ref>{{cite journal |last1=Romem |first1=Yoram |title=The Social Construction of Expert Systems |journal=Human Systems Management |date=2007 |volume=26 |issue=4 |pages=291–309 |doi=10.3233/HSM-2007-26406 |url=https://www.researchgate.net/publication/228630533}}</ref>

In the 1990s and beyond, the term ''expert system'' and the idea of a standalone AI system mostly dropped from the IT lexicon. There are two interpretations of this. One is that "expert systems failed": the IT world moved on because expert systems did not deliver on their over hyped promise.<ref>{{Cite web |url=http://www.ainewsletter.com/newsletters/aix_0501.htm#w |title=AI Expert Newsletter: W is for Winter |access-date=2013-11-29 |archive-url=https://web.archive.org/web/20131109201636/http://www.ainewsletter.com/newsletters/aix_0501.htm#w |archive-date=2013-11-09 |url-status=dead}}</ref><ref>{{cite journal |last1=Leith |first1=P. |date=2010 |url=http://ejlt.org//article/view/14/1 |title=The rise and fall of the legal expert system |journal=European Journal of Law and Technology |volume=1 |issue=1 |access-date=2020-01-24 |url-status=dead |archive-date=2016-03-04 |archive-url=https://web.archive.org/web/20160304124042/http://ejlt.org//article/view/14/1}}</ref> The other is the mirror opposite, that expert systems were simply victims of their success: as IT professionals grasped concepts such as rule engines, such tools migrated from being standalone tools for developing special purpose ''expert'' systems, to being one of many standard tools.<ref>{{cite journal |last1=Haskin |first1=David |date=January 16, 2003 |title=Years After Hype, 'Expert Systems' Paying Off for Some |journal=Datamation |url=http://www.datamation.com/netsys/article.php/1570851/Years-After-Hype-Expert-Systems-Paying-Off-For-Some.htm |access-date=29 November 2013 |archive-date=3 December 2013 |archive-url=https://web.archive.org/web/20131203013322/http://www.datamation.com/netsys/article.php/1570851/Years-After-Hype-Expert-Systems-Paying-Off-For-Some.htm |url-status=live }}</ref> Other researchers suggest that Expert Systems caused inter-company power struggles when the IT organization lost its exclusivity in software modifications to users or Knowledge Engineers.<ref>{{cite journal |last1=Romem |first1=Yoram |title=The Social Construction of Expert Systems |journal=Human Systems Management |date=2007 |volume=26 |issue=4 |pages=291–309 |doi=10.3233/HSM-2007-26406 |url=https://www.researchgate.net/publication/228630533}}</ref>

In the first decade of the 2000s, there was a "resurrection" for the technology, while using the term ''[[rule-based system]]s'', with significant success stories and adoption.<ref>{{cite news |last1=Voelker |first1=Michael P. |date=October 18, 2005 |url=https://www.informationweek.com/information-management/business-makes-the-rules |title=Business Makes the Rules |work=Information Week}}</ref> Many of the leading major business application suite vendors (such as [[SAP (software)|SAP]], [[Siebel Systems|Siebel]], and [[Oracle Corporation|Oracle]]) integrated expert system abilities into their suite of products as a way to specify business logic. Rule engines are no longer simply for defining the rules an expert would use but for any type of complex, volatile, and critical business logic; they often go hand in hand with business process automation and integration environments.<ref>{{cite web |last=SAP News Desk |title=SAP News Desk IntelliCorp Announces Participation in SAP EcoHub |url=http://laszlo.sys-con.com/node/946452 |work=laszlo.sys-con.com |publisher=LaszloTrack |access-date=29 November 2013 |archive-date=3 December 2013 |archive-url=https://web.archive.org/web/20131203002523/http://laszlo.sys-con.com/node/946452 |url-status=dead}}</ref><ref>{{cite web |last=Pegasystems |title=Smart BPM Requires Smart Business Rules |url=http://www.pega.com/business-rules |work=pega.com |access-date=29 November 2013}}</ref><ref>{{cite book |last1=Zhao |first1=Kai |first2=Shi |last2=Ying |first3=Linlin |last3=Zhang |first4=Luokai |last4=Hu |title=2010 International Conference on Future Information Technology and Management Engineering |chapter=Achieving business process and business rules integration using SPL |date=9–10 Oct 2010 |volume=2 |pages=329–332 |publisher=IEEE |doi=10.1109/fitme.2010.5656297 |isbn=978-1-4244-9087-5}}</ref>

In the first decade of the 2000s, there was a "resurrection" for the technology, while using the term ''[[rule-based system]]s'', with significant success stories and adoption.<ref>{{cite news |last1=Voelker |first1=Michael P. |date=October 18, 2005 |url=https://www.informationweek.com/information-management/business-makes-the-rules |title=Business Makes the Rules |work=Information Week |archive-date=December 23, 2022 |access-date=December 23, 2022 |archive-url=https://web.archive.org/web/20221223174411/https://www.informationweek.com/information-management/business-makes-the-rules |url-status=live }}</ref> Many of the leading major business application suite vendors (such as [[SAP (software)|SAP]], [[Siebel Systems|Siebel]], and [[Oracle Corporation|Oracle]]) integrated expert system abilities into their suite of products as a way to specify business logic. Rule engines are no longer simply for defining the rules an expert would use but for any type of complex, volatile, and critical business logic; they often go hand in hand with business process automation and integration environments.<ref>{{cite web |last=SAP News Desk |title=SAP News Desk IntelliCorp Announces Participation in SAP EcoHub |url=http://laszlo.sys-con.com/node/946452 |work=laszlo.sys-con.com |publisher=LaszloTrack |access-date=29 November 2013 |archive-date=3 December 2013 |archive-url=https://web.archive.org/web/20131203002523/http://laszlo.sys-con.com/node/946452 |url-status=dead}}</ref><ref>{{cite web |last=Pegasystems |title=Smart BPM Requires Smart Business Rules |url=http://www.pega.com/business-rules |work=pega.com |access-date=29 November 2013 |archive-date=9 November 2013 |archive-url=https://web.archive.org/web/20131109110005/http://www.pega.com/business-rules |url-status=live }}</ref><ref>{{cite book |last1=Zhao |first1=Kai |first2=Shi |last2=Ying |first3=Linlin |last3=Zhang |first4=Luokai |last4=Hu |title=2010 International Conference on Future Information Technology and Management Engineering |chapter=Achieving business process and business rules integration using SPL |date=9–10 Oct 2010 |volume=2 |pages=329–332 |publisher=IEEE |doi=10.1109/fitme.2010.5656297 |isbn=978-1-4244-9087-5}}</ref>

=== Current approaches to expert systems ===

The limits of prior ~~type~~ of expert systems prompted researchers to develop new types of approaches. They have developed more efficient, flexible, and powerful methods to simulate the human decision-making process. Some of the approaches that researchers have developed are based on new methods of artificial intelligence (AI), and in particular in [[machine learning]] and [[data mining]] approaches with a feedback mechanism.<ref>{{Cite journal |last1=Chung |first1=Junyoung |last2=Gulcehre |first2=Caglar |last3=Cho |first3=Kyunghyun |last4=Bengio |first4=Yoshua |date=2015-06-01 |title=Gated Feedback Recurrent Neural Networks |url=http://proceedings.mlr.press/v37/chung15.html |journal=International Conference on Machine Learning |language=en |publisher=PMLR |pages=2067–2075 |arxiv=1502.02367}}</ref>{{failed verification|reason=The article doesn't mention expert systems at all.|date=May 2024}} [[Recurrent neural network]]s often take advantage of such mechanisms. Related is the discussion on the disadvantages section.

The limits of prior types of expert systems prompted researchers to develop new types of approaches. They have developed more efficient, flexible, and powerful methods to simulate the human decision-making process. Some of the approaches that researchers have developed are based on new methods of artificial intelligence (AI), and in particular in [[machine learning]] and [[data mining]] approaches with a feedback mechanism.<ref>{{Cite journal |last1=Chung |first1=Junyoung |last2=Gulcehre |first2=Caglar |last3=Cho |first3=Kyunghyun |last4=Bengio |first4=Yoshua |date=2015-06-01 |title=Gated Feedback Recurrent Neural Networks |url=http://proceedings.mlr.press/v37/chung15.html |journal=International Conference on Machine Learning |language=en |publisher=PMLR |pages=2067–2075 |arxiv=1502.02367 |archive-date=2021-05-24 |access-date=2021-05-24 |archive-url=https://web.archive.org/web/20210524143259/http://proceedings.mlr.press/v37/chung15.html |url-status=live }}</ref>{{failed verification|reason=The article doesn't mention expert systems at all.|date=May 2024}} [[Recurrent neural network]]s often take advantage of such mechanisms. Related is the discussion on the disadvantages section.

Modern systems can incorporate new knowledge more easily and thus update themselves easily. Such systems can generalize from existing knowledge better and deal with vast amounts of complex data. Related is the subject of [[big data]] here. Sometimes these types of expert systems are called "intelligent systems."<ref name="CADsurvey"/>

Line 46:

Line 48:

== Software architecture ==

[[File:BackwardChaining David C England 1990 p21.gif|thumb|450px|Illustrating example of ''backward chaining'' from a 1990 Master's Thesis<ref>{{cite thesis |type=MSc |url=https://commons.wikimedia.org/wiki/File:An_expert_system_for_the_management_of_hazardous_materials_at_a_Naval_Supply_Center_(IA_anexpertsystemfo1094530640).pdf |last1=England |first1=David C. |date=June 1990 |title=An Expert System for the Management of Hazardous Materials at a Naval Supply Center |publisher=Naval Postgraduate School Monterey/CA |page=21}}</ref>]]

[[File:BackwardChaining David C England 1990 p21.gif|thumb|450px|Illustrating example of ''backward chaining'' from a 1990 Master's Thesis<ref>{{cite thesis |type=MSc |url=https://commons.wikimedia.org/wiki/File:An_expert_system_for_the_management_of_hazardous_materials_at_a_Naval_Supply_Center_(IA_anexpertsystemfo1094530640).pdf |last1=England |first1=David C. |date=June 1990 |title=An Expert System for the Management of Hazardous Materials at a Naval Supply Center |publisher=Naval Postgraduate School Monterey/CA |page=21 |archive-date=2021-10-18 |access-date=2020-09-01 |archive-url=https://web.archive.org/web/20211018080757/https://commons.wikimedia.org/wiki/File:An_expert_system_for_the_management_of_hazardous_materials_at_a_Naval_Supply_Center_(IA_anexpertsystemfo1094530640).pdf |url-status=live }}</ref>]]

An expert system is an example of a [[knowledge-based system]]. Expert systems were the first commercial systems to use a knowledge-based architecture. In general view, an expert system includes the following components: a [[knowledge base]], an [[inference engine]], an explanation facility, a knowledge acquisition facility, and a user interface.<ref name=":0">{{Cite journal |last=Kiryanov |first=Denis Aleksandrovich|date=2021-12-21 |title=Hybrid categorical expert system for use in content aggregation |journal=Software Systems and Computational Methods |issue=4 |pages=1–22|doi=10.7256/2454-0714.2021.4.37019 |s2cid=245498498 |issn=2454-0714|doi-access=free}}</ref><ref>{{cite web |last=Smith |first=Reid|date=May 8, 1985 |title=Knowledge-Based Systems Concepts, Techniques, Examples|url=http://www.reidgsmith.com/Knowledge-Based_Systems_-_Concepts_Techniques_Examples_08-May-1985.pdf|access-date=9 November 2013|website=Reid G. Smith}}</ref>

An expert system is an example of a [[knowledge-based system]]. Expert systems were the first commercial systems to use a knowledge-based architecture. In general view, an expert system includes the following components: a [[knowledge base]], an [[inference engine]], an explanation facility, a knowledge acquisition facility, and a user interface.<ref name=":0">{{Cite journal |last=Kiryanov |first=Denis Aleksandrovich|date=2021-12-21 |title=Hybrid categorical expert system for use in content aggregation |journal=Software Systems and Computational Methods |issue=4 |pages=1–22|doi=10.7256/2454-0714.2021.4.37019 |s2cid=245498498 |issn=2454-0714|doi-access=free}}</ref><ref>{{cite web|last=Smith|first=Reid|date=May 8, 1985|title=Knowledge-Based Systems Concepts, Techniques, Examples|url=http://www.reidgsmith.com/Knowledge-Based_Systems_-_Concepts_Techniques_Examples_08-May-1985.pdf|access-date=9 November 2013|website=Reid G. Smith|archive-date=10 February 2023|archive-url=https://web.archive.org/web/20230210113921/http://www.reidgsmith.com/Knowledge-Based_Systems_-_Concepts_Techniques_Examples_08-May-1985.pdf|url-status=live}}</ref>

The knowledge base represents facts about the world. In early expert systems such as Mycin and Dendral, these facts were represented mainly as flat assertions about variables. In later expert systems developed with commercial shells, the knowledge base took on more structure and used concepts from [[object-oriented programming]]. The world was represented as [[Class (~~computer~~ programming)|classes, subclasses]], and [[Instance (computer science)|instances]] and [[Assertion (software development)|assertions]] were replaced by values of object instances. The rules worked by querying and asserting values of the objects.

The knowledge base represents facts about the world. In early expert systems such as Mycin and Dendral, these facts were represented mainly as flat assertions about variables. In later expert systems developed with commercial shells, the knowledge base took on more structure and used concepts from [[object-oriented programming]]. The world was represented as [[Class (programming)|classes, subclasses]], and [[Instance (computer science)|instances]] and [[Assertion (software development)|assertions]] were replaced by values of object instances. The rules worked by querying and asserting values of the objects.

The inference engine is an [[automated reasoning system]] that evaluates the current state of the knowledge-base, applies relevant rules, and then asserts new knowledge into the knowledge base. The inference engine may also include abilities for explanation, so that it can explain to a user the chain of reasoning used to arrive at a particular conclusion by tracing back over the firing of rules that resulted in the assertion.<ref name="Hayes-Roth 1983">{{cite book |last1=Hayes-Roth |first1=Frederick |last2=Waterman |first2=Donald |last3=Lenat |first3=Douglas |author3-link=Douglas Lenat |year=1983 |title=Building Expert Systems |publisher=Addison-Wesley |isbn=978-0-201-10686-2 |url=https://archive.org/details/buildingexpertsy00temd}}</ref>

Line 61:

Line 63:

Backward chaining is a bit less straight forward. In backward chaining the system looks at possible conclusions and works backward to see if they might be true. So if the system was trying to determine if Mortal(Socrates) is true it would find R1 and query the knowledge base to see if Man(Socrates) is true. One of the early innovations of expert systems shells was to integrate inference engines with a user interface. This could be especially powerful with backward chaining. If the system needs to know a particular fact but does not, then it can simply generate an input screen and ask the user if the information is known. So in this example, it could use R1 to ask the user if Socrates was a Man and then use that new information accordingly.

The use of rules to explicitly represent knowledge also enabled explanation abilities. In the simple example above if the system had used R1 to assert that Socrates was Mortal and a user wished to understand why Socrates was mortal they could query the system and the system would look back at the rules which fired to cause the assertion and present those rules to the user as an explanation. In English, if the user asked "Why is Socrates Mortal?" the system would reply "Because all men are mortal and Socrates is a man". A significant area for research was the generation of explanations from the knowledge base in natural English rather than simply by showing the more formal but less intuitive rules.<ref>[http://www.ccis2k.org/iajit/PDF/vol.4,no.1/9-Nabil.pdf Nabil Arman], Polytechnic University of Palestine, January 2007, Fault Detection in Dynamic Rule Bases Using Spanning Trees and Disjoin Sets: ""</ref>

The use of rules to explicitly represent knowledge also enabled explanation abilities. In the simple example above if the system had used R1 to assert that Socrates was Mortal and a user wished to understand why Socrates was mortal they could query the system and the system would look back at the rules which fired to cause the assertion and present those rules to the user as an explanation. In English, if the user asked "Why is Socrates Mortal?" the system would reply "Because all men are mortal and Socrates is a man". A significant area for research was the generation of explanations from the knowledge base in natural English rather than simply by showing the more formal but less intuitive rules.<ref>[http://www.ccis2k.org/iajit/PDF/vol.4,no.1/9-Nabil.pdf Nabil Arman] {{Webarchive|url=https://web.archive.org/web/20120425141837/http://www.ccis2k.org/iajit/PDF/vol.4,no.1/9-Nabil.pdf |date=2012-04-25 }}, Polytechnic University of Palestine, January 2007, Fault Detection in Dynamic Rule Bases Using Spanning Trees and Disjoin Sets: ""</ref>

As expert systems evolved, many new techniques were incorporated into various types of inference engines.<ref>{{cite journal |last=Mettrey |first=William |title=An Assessment of Tools for Building Large Knowledge-Based Systems |journal=AI Magazine |year=1987 |volume=8 |issue=4|url=http://www.aaai.org/ojs/index.php/aimagazine/article/viewArticle/625|access-date=2013-11-29 |archive-url=https://web.archive.org/web/20131110022104/http://www.aaai.org/ojs/index.php/aimagazine/article/viewArticle/625 |archive-date=2013-11-10|url-status=dead}}</ref> Some of the most important of these were:

Line 67:

Line 69:

* Hypothetical reasoning. In this, the knowledge base can be divided up into many possible views, a.k.a. worlds. This allows the inference engine to explore multiple possibilities in parallel. For example, the system may want to explore the consequences of both assertions, what will be true if Socrates is a Man and what will be true if he is not?

* Uncertainty systems. One of the first extensions of simply using rules to represent knowledge was also to associate a probability with each rule. So, not to assert that Socrates is mortal, but to assert Socrates ''may'' be mortal with some probability value. Simple probabilities were extended in some systems with sophisticated mechanisms for uncertain reasoning, such as [[fuzzy logic]], and combination of probabilities.

* [[Ontology (information science)|Ontology]] classification. With the addition of object classes to the knowledge base, a new type of reasoning was possible. Along with reasoning simply about object values, the system could also reason about object structures. In this simple example, Man can represent an object class and R1 can be redefined as a rule that defines the class of all men. These types of special purpose inference engines are termed [[Deductive classifier|classifiers]]. Although they were not highly used in expert systems, classifiers are very powerful for unstructured volatile domains, and are a key technology for the Internet and the emerging [[Semantic Web]].<ref>{{cite journal |last1=MacGregor |first1=Robert |date=June 1991 |title=Using a description classifier to enhance knowledge representation |journal=IEEE Expert |volume=6 |issue=3 |doi=10.1109/64.87683 |pages=41–46 |s2cid=29575443}}</ref><ref>{{cite journal |last1=Berners-Lee |first1=Tim |last2=Hendler |first2=James |last3=Lassila |first3=Ora |date=May 17, 2001 |title=The Semantic Web A new form of Web content that is meaningful to computers will unleash a revolution of new possibilities |journal=Scientific American |url=http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |doi=10.1038/scientificamerican0501-34 |volume=284 |issue=5 |pages=34–43 |url-status=dead |archive-url=https://web.archive.org/web/20130424071228/http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |archive-date=April 24, 2013 |url-access=subscription}}</ref>

* [[Ontology (information science)|Ontology]] classification. With the addition of object classes to the knowledge base, a new type of reasoning was possible. Along with reasoning simply about object values, the system could also reason about object structures. In this simple example, Man can represent an object class and R1 can be redefined as a rule that defines the class of all men. These types of special purpose inference engines are termed [[Deductive classifier|classifiers]]. Although they were not highly used in expert systems, classifiers are very powerful for unstructured volatile domains, and are a key technology for the Internet and the emerging [[Semantic Web]].<ref>{{cite journal |last1=MacGregor |first1=Robert |date=June 1991 |title=Using a description classifier to enhance knowledge representation |journal=IEEE Expert |volume=6 |issue=3 |doi=10.1109/64.87683 |pages=41–46 |bibcode=1991IExp....6...41M |s2cid=29575443}}</ref><ref>{{cite journal |last1=Berners-Lee |first1=Tim |last2=Hendler |first2=James |last3=Lassila |first3=Ora |date=May 17, 2001 |title=The Semantic Web A new form of Web content that is meaningful to computers will unleash a revolution of new possibilities |journal=Scientific American |url=http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |doi=10.1038/scientificamerican0501-34 |volume=284 |issue=5 |pages=34–43 |url-status=dead |archive-url=https://web.archive.org/web/20130424071228/http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |archive-date=April 24, 2013 |url-access=subscription}}</ref>

== Advantages ==

Line 90:

Line 92:

Another major challenge of expert systems emerges when the size of the knowledge base increases. This causes the processing complexity to increase. For instance, when an expert system with 100 million rules was envisioned as the ultimate expert system, it became obvious that such system would be too complex and it would face too many computational problems.<ref>{{cite book |last1=Lenat |first1=Douglas |author1-link=Douglas Lenat |chapter=On the thresholds of knowledge |title=Foundations Of Artificial Intelligence |editor1-last=Kirsh |editor1-first=David |publisher=MIT Press |pages=185–250 |year=1992}}</ref> An inference engine would have to be able to process huge numbers of rules to reach a decision.

How to verify that decision rules are consistent with each other is also a challenge when there are too many rules. Usually such problem leads to a [[satisfiability]] (SAT) formulation.<ref>{{cite conference |vauthors=Bezem M |work=9th International Conference on Automated Deduction |~~title~~=Consistency of rule-based expert systems |volume=310 |date=1988 |pages=151–161 |doi=10.1007/BFb0012830 |isbn=3-540-19343-X |series=Lecture Notes in Computer Science |url=https://ir.cwi.nl/pub/6113|url-access=subscription }}</ref> This is a well-known NP-complete problem [[Boolean satisfiability problem]]. If we assume only [[binary data|binary variables]], say n of them, and then the corresponding search space is of size 2<math>^{n}</math>. Thus, the search space can grow exponentially.

How to verify that decision rules are consistent with each other is also a challenge when there are too many rules. Usually such problem leads to a [[satisfiability]] (SAT) formulation.<ref>{{cite conference |vauthors=Bezem M |work=9th International Conference on Automated Deduction |chapter=Consistency of rule-based expert systems |volume=310 |date=1988 |pages=151–161 |doi=10.1007/BFb0012830 |isbn=3-540-19343-X |series=Lecture Notes in Computer Science |chapter-url=https://ir.cwi.nl/pub/6113 |chapter-url-access=subscription |title=Archived copy |access-date=2020-08-30 |archive-date=2020-10-23 |archive-url=https://web.archive.org/web/20201023223526/https://ir.cwi.nl/pub/6113 |url-status=live }}</ref> This is a well-known NP-complete problem [[Boolean satisfiability problem]]. If we assume only [[binary data|binary variables]], say n of them, and then the corresponding search space is of size 2<math>^{n}</math>. Thus, the search space can grow exponentially.

There are also questions on how to prioritize the use of the rules to operate more efficiently, or how to resolve ambiguities (for instance, if there are too many else-if sub-structures within one rule) and so on.<ref>{{cite journal |vauthors=Mak B, Schmitt BH, and Lyytinen K |title=User participation in knowledge update of expert systems |journal=Information & Management |date=1997 |volume=32 |issue=2 |pages=55–63 |doi=10.1016/S0378-7206(96)00010-9 |doi-access=free}}</ref>

Other problems are related to the [[overfitting]] and [[overgeneralization]] effects when using known facts and trying to generalize to other cases not described explicitly in the knowledge base. Such problems exist with methods that employ machine learning approaches too.<ref>{{cite book |vauthors=Pham HN, Triantaphyllou E |title=Soft Computing for Knowledge Discovery and Data Mining |chapter=The Impact of Overfitting and Overgeneralization on the Classification Accuracy in Data Mining |date=2008 |pages=391–431 |doi=10.1007/978-0-387-69935-6_16 |isbn=978-0-387-69934-9 |s2cid=12628921 |chapter-url=https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=4334&context=gradschool_dissertations}}</ref><ref>{{cite journal |vauthors=Pham HN, Triantaphyllou E |title=Prediction of diabetes by employing a new data mining approach which balances fitting and generalization |journal=Computer and Inf. Science G |date=2008 |pages=11–26}}</ref>

Other problems are related to the [[overfitting]] and [[overgeneralization]] effects when using known facts and trying to generalize to other cases not described explicitly in the knowledge base. Such problems exist with methods that employ machine learning approaches too.<ref>{{cite book |vauthors=Pham HN, Triantaphyllou E |title=Soft Computing for Knowledge Discovery and Data Mining |chapter=The Impact of Overfitting and Overgeneralization on the Classification Accuracy in Data Mining |date=2008 |pages=391–431 |doi=10.1007/978-0-387-69935-6_16 |isbn=978-0-387-69934-9 |s2cid=12628921 |chapter-url=https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=4334&context=gradschool_dissertations |archive-date=2023-07-05 |access-date=2023-07-23 |archive-url=https://web.archive.org/web/20230705040238/https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=4334&context=gradschool_dissertations |url-status=live }}</ref><ref>{{cite journal |vauthors=Pham HN, Triantaphyllou E |title=Prediction of diabetes by employing a new data mining approach which balances fitting and generalization |journal=Computer and Inf. Science G |date=2008 |pages=11–26}}</ref>

Another problem related to the knowledge base is how to make updates of its knowledge quickly and effectively.<ref>{{cite journal |vauthors=Shan N, and Ziarko W |title=Data-based acquisition and incremental modification of classification rules |journal=Computational Intelligence |date=1995 |volume=11 |issue=2 |pages=357–370 |doi=10.1111/j.1467-8640.1995.tb00038.x |s2cid=38974914}}</ref><ref>{{cite journal |vauthors=Coats PK |title=Why expert systems fail |journal=Financial Management |volume=17 |issue=3 |date=1988 |pages=77–86 |jstor=3666074}}</ref><ref>{{cite journal |vauthors=Hendriks PH, and Vriens DJ |title=Knowledge-based systems and knowledge management: friends or foes? |journal=Information & Management |date=1999 |volume=35 |issue=2 |pages=113–125 |doi=10.1016/S0378-7206(98)00080-9 |url=https://repository.ubn.ru.nl/handle/2066/240554|hdl=2066/240554 |hdl-access=free }}</ref> Also how to add a new piece of knowledge (i.e., where to add it among many rules) is challenging. Modern approaches that rely on machine learning methods are easier in this regard.{{Citation needed|date=October 2019}}

Another problem related to the knowledge base is how to make updates of its knowledge quickly and effectively.<ref>{{cite journal |vauthors=Shan N, and Ziarko W |title=Data-based acquisition and incremental modification of classification rules |journal=Computational Intelligence |date=1995 |volume=11 |issue=2 |pages=357–370 |doi=10.1111/j.1467-8640.1995.tb00038.x |s2cid=38974914}}</ref><ref>{{cite journal |vauthors=Coats PK |title=Why expert systems fail |journal=Financial Management |volume=17 |issue=3 |date=1988 |pages=77–86 |jstor=3666074}}</ref><ref>{{cite journal |vauthors=Hendriks PH, and Vriens DJ |title=Knowledge-based systems and knowledge management: friends or foes? |journal=Information & Management |date=1999 |volume=35 |issue=2 |pages=113–125 |doi=10.1016/S0378-7206(98)00080-9 |url=https://repository.ubn.ru.nl/handle/2066/240554 |hdl=2066/240554 |hdl-access=free |archive-date=2024-04-19 |access-date=2024-02-20 |archive-url=https://web.archive.org/web/20240419185339/https://repository.ubn.ru.nl/handle/2066/240554 |url-status=live }}</ref> Also how to add a new piece of knowledge (i.e., where to add it among many rules) is challenging. Modern approaches that rely on machine learning methods are easier in this regard.{{Citation needed|date=October 2019}}

Because of the above challenges, it became clear that new approaches to AI were required instead of rule-based technologies. These new approaches are based on the use of machine learning techniques, along with the use of feedback mechanisms.<ref name="CADsurvey"/>

The key challenges that expert systems in medicine (if one considers computer-aided diagnostic systems as modern expert systems), and perhaps in other application domains, include issues related to aspects such as: big data, existing regulations, healthcare practice, various algorithmic issues, and system assessment.<ref>{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=The Seven Key Challenges for the Future of Computer-Aided Diagnosis in Medicine |journal=International Journal of Medical Informatics |volume=129 |pages=413–422 |date=2019 |doi=10.1016/j.ijmedinf.2019.06.017 |pmid=31445285 |s2cid=198287435}}</ref>

The key challenges that expert systems in medicine (if one considers computer-aided diagnostic systems as modern expert systems), and perhaps in other application domains, include issues related to aspects such as: big data, existing regulations, healthcare practice, various algorithmic issues, and system assessment.<ref>{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=The Seven Key Challenges for the Future of Computer-Aided Diagnosis in Medicine |journal=International Journal of Medical Informatics |volume=129 |pages=413–422 |date=2019 |doi=10.1016/j.ijmedinf.2019.06.017 |pmid=31445285 |s2cid=198287435 |url=https://repository.lsu.edu/eecs_pubs/1347 |archive-date=2025-08-13 |access-date=2026-06-01 |archive-url=https://web.archive.org/web/20250813104551/https://repository.lsu.edu/eecs_pubs/1347/ |url-status=live }}</ref>

Finally, the following disadvantages of using expert systems can be summarized:<ref name=":0"/>

Line 120:

Line 122:

| Prediction || Inferring likely consequences of given situations || Preterm Birth Risk Assessment<ref>{{cite journal |last1=Woolery |first1=L.K. |last2=Grzymala-Busse |first2=J. |year=1994 |title=Machine learning for an expert system to predict preterm birth risk |journal=Journal of the American Medical Informatics Association |volume=1 |issue=6 |pages=439–446 |pmc=116227 |pmid=7850569 |doi=10.1136/jamia.1994.95153433}}</ref>

|-

| Diagnosis || Inferring system malfunctions from observables || [[CADUCEUS (expert system)|CADUCEUS]], [[MYCIN]], PUFF, Mistral,<ref name="Mistral">{{cite journal |last1=Salvaneschi |first1=Paolo |first2=Mauro |last2=Cadei |first3=Marco |last3=Lazzari |year=1996 |title=Applying AI to structural safety monitoring and evaluation |journal=IEEE Expert |volume=11 |issue=4 |pages=24–34 |url=http://www.computer.org/csdl/mags/ex/1996/04/x4024-abs.html|access-date=5 March 2014 |doi=10.1109/64.511774 |url-access=subscription}}</ref> Eydenet,<ref name="Eydenet"/> Kaleidos,<ref name="Kaleidos"/> GARVAN-ES1<ref name="GARVAN">{{cite journal |author1=K. Horn |author2=L. Lazarus |author3=P. Compton |author4=J.R. Quinlan |title=An expert system for the interpretation of thyroid assays in a clinical laboratory |journal=Aust Comput J |year=1985 |volume=17 |pages=7–11}}</ref><ref name="Buchanan">{{cite journal |author=Buchanan, B. |title=Expert systems: working systems and the research literature |journal=Expert Systems |year=1986 |volume=3 |issue=1 |pages=32–51 |doi= 10.1111/j.1468-0394.1986.tb00192.x}}</ref><ref name="Maintenance">{{cite journal |author1=P. Compton |author2=K. Horn |author3=R. Quinlan |author4=L. Lazarus |author5=K. Ho |title=Maintaining an Expert System |journal=Proceedings of the Fourth Australian Conference on the Applications of Expert Systems |year=1988}}</ref>

| Diagnosis || Inferring system malfunctions from observables || [[CADUCEUS (expert system)|CADUCEUS]], [[MYCIN]], PUFF, Mistral,<ref name="Mistral">{{cite journal |last1=Salvaneschi |first1=Paolo |first2=Mauro |last2=Cadei |first3=Marco |last3=Lazzari |year=1996 |title=Applying AI to structural safety monitoring and evaluation |journal=IEEE Expert |volume=11 |issue=4 |pages=24–34 |url=http://www.computer.org/csdl/mags/ex/1996/04/x4024-abs.html |access-date=5 March 2014 |doi=10.1109/64.511774 |bibcode=1996IExp...11...24S |url-access=subscription |archive-date=5 March 2014 |archive-url=https://web.archive.org/web/20140305053153/http://www.computer.org/csdl/mags/ex/1996/04/x4024-abs.html |url-status=live }}</ref> Eydenet,<ref name="Eydenet"/> Kaleidos,<ref name="Kaleidos"/> GARVAN-ES1<ref name="GARVAN">{{cite journal |author1=K. Horn |author2=L. Lazarus |author3=P. Compton |author4=J.R. Quinlan |title=An expert system for the interpretation of thyroid assays in a clinical laboratory |journal=Aust Comput J |year=1985 |volume=17 |pages=7–11}}</ref><ref name="Buchanan">{{cite journal |author=Buchanan, B. |title=Expert systems: working systems and the research literature |journal=Expert Systems |year=1986 |volume=3 |issue=1 |pages=32–51 |doi= 10.1111/j.1468-0394.1986.tb00192.x}}</ref><ref name="Maintenance">{{cite journal |author1=P. Compton |author2=K. Horn |author3=R. Quinlan |author4=L. Lazarus |author5=K. Ho |title=Maintaining an Expert System |journal=Proceedings of the Fourth Australian Conference on the Applications of Expert Systems |year=1988}}</ref>

|-

| Design || Configuring objects under constraints || [[Dendral]], [[Expert systems for mortgages|Mortgage Loan Advisor]], [[Xcon|R1]] (DEC VAX Configuration), SID (DEC [[VAX 9000]] [[central processing unit|CPU]])

| Design || Configuring objects under constraints || [[Dendral]], [[Expert systems for mortgages|Mortgage Loan Advisor]], [[Xcon|R1]] (DEC VAX Configuration), SID (DEC [[VAX 9000]] [[central processing unit|CPU]]), Database Design Advisor<ref>{{cite conference

|chapter=Design and implementation of a database design aid using VP-Expert

|last1=Song

|first1=Il-Yeol

|last2=Strum

|first2=S. D.

|last3=Medsker

|first3=Carl J.

|title=[1991] Proceedings of the IEEE/ACM International Conference on Developing and Managing Expert System Programs

|year=1991

|pages=15–23

|publisher=IEEE

|doi=10.1109/DMESP.1991.171764

|isbn=0-8186-2250-4

|chapter-url=https://www.computer.org/csdl/proceedings-article/dmesp/1991/00171764/12OmNrkT7EE

|access-date=2025-09-28

|archive-date=2024-12-15

|archive-url=https://web.archive.org/web/20241215023737/https://www.computer.org/csdl/proceedings-article/dmesp/1991/00171764/12OmNrkT7EE

|url-status=live

}}</ref>

|-

| [[Automated planning and scheduling|Planning]] || Designing actions || Mission Planning for Autonomous Underwater Vehicle<ref>{{cite book |last=Kwak |first=S. H. |title=Symposium on Autonomous Underwater Vehicle Technology |chapter=A mission planning expert system for an autonomous underwater vehicle |year=1990 |pages=123–128|doi=10.1109/AUV.1990.110446 |s2cid=60476847}}</ref>

|-

| Monitoring || Comparing observations to plan vulnerabilities || REACTOR<ref>{{cite journal |last=Nelson |first=W. R. |title=REACTOR: An Expert System for Diagnosis and Treatment of Nuclear Reactors |url=https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4daf79217bb76482943f07b5138104a45f85c076 |year=1982 |journal=AAAI |volume=82: Proceedings of the Second AAAI Conference on Artificial Intelligence |pages=296–301}}</ref>

| Monitoring || Comparing observations to plan vulnerabilities || REACTOR<ref>{{cite journal |last=Nelson |first=W. R. |title=REACTOR: An Expert System for Diagnosis and Treatment of Nuclear Reactors |url=https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4daf79217bb76482943f07b5138104a45f85c076 |year=1982 |journal=AAAI |volume=82: Proceedings of the Second AAAI Conference on Artificial Intelligence |pages=296–301 |archive-date=2023-03-11 |access-date=2023-03-11 |archive-url=https://web.archive.org/web/20230311174105/https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4daf79217bb76482943f07b5138104a45f85c076 |url-status=live }}</ref>

|-

| Debugging || Providing incremental solutions for complex problems || SAINT, MATHLAB, MACSYMA

Line 134:

Line 156:

| Instruction || Diagnosing, assessing, and correcting student behaviour || SMH.PAL,<ref name="smhpal"/> Intelligent Clinical Training,<ref>{{cite journal |last1=Haddawy |first1=P |first2=S. |last2=Suebnukarn |year=2010 |title=Intelligent Clinical Training Systems |journal=Methods Inf Med |volume=49 |issue=4 |pages=388–9 |citeseerx=10.1.1.172.60 |pmid=20686730|doi=10.1055/s-0038-1625342 |s2cid=11903941}}</ref> STEAMER<ref>{{cite journal |last1=Hollan |first1=J. |first2=E. |last2=Hutchins |first3=L. |last3=Weitzman |year=1984 |title=STEAMER: An interactive inspectable simulation-based training system |journal=AI Magazine}}</ref>

|-

| Control || Interpreting, predicting, repairing, and monitoring system behaviors || Real Time Process Control,<ref>{{cite journal |last=Stanley |first=G.M. |title=Experience Using Knowledge-Based Reasoning in Real Time Process Control |journal=Plenary Paper Presented at: International Federation of Automatic Control (IFAC) Symposium on Compute R Aided Design in Control Systems|date=July 15–17, 1991|url=http://www.gregstanleyandassociates.com/whitepapers/IFAC91objectPaper.pdf|access-date=3 December 2013}}</ref> Space Shuttle Mission Control,<ref>{{cite journal |last1=Rasmussen |first1=Arthur |first2=John F. |last2=Muratore |first3=Troy A. |last3=Heindel |title=The INCO Expert System Project: CLIPS in Shuttle mission control |journal=NTRS|date=February 1990|url=https://www.researchgate.net/publication/4702412|access-date=30 November 2013}}</ref> Smart Autoclave Cure of Composites<ref>{{Cite book |author1=Ciriscioli, P. R. |author2=G. S. Springer |title="Smart Autoclave Cure of Composites" |year=1990 |isbn=9781003209010}}</ref>

|}

Line 147:

Line 170:

GARVAN-ES1 was a medical expert system, developed at the [[Garvan Institute of Medical Research]], that provided automated clinical diagnostic comments on endocrine reports from a pathology laboratory. It was one of the first medical expert systems to go into routine clinical use internationally<ref name="Buchanan"/> and the first expert system to be used for diagnosis daily in Australia.<ref name="Catlett">{{cite journal |author=Catlett, J. |title=Expert systems, the risks and rewards |journal=Hub Information Technology |year=1990 |volume=2 |issue=7 |pages=20–26}}</ref> The system was written in "C" and ran on a PDP-11 in 64K of memory. It had 661 rules that were compiled; not interpreted.

Mistral<ref name="Mistral"/> is an expert system to monitor dam safety, developed in the 1990s by Ismes (Italy). It gets data from an automatic monitoring system and performs a diagnosis of the state of the dam. Its first copy, installed in 1992 on the [[Ridracoli]] Dam (Italy~~), is still operational 24/7/365~~. It has been installed on several dams in Italy and abroad (e.g., [[Itaipu Dam]] in Brazil), and on landslide sites under the name of Eydenet,<ref name="Eydenet">{{cite journal |last1=Lazzari |first1=Marco |first2=Paolo |last2=Salvaneschi |title=Embedding a geographic information system in a decision support system for landslide hazard monitoring |journal=International Journal of Natural Hazards |year=1999 |volume=20 |issue=2–3 |pages=185–195|url=http://dinamico2.unibg.it/lazzari/doc/embedding-authors-copy.pdf|doi=10.1023/A:1008187024768 |bibcode=1999NatHa..20..185L |s2cid=1746570}}</ref> and on monuments under the name of Kaleidos.<ref name="Kaleidos">{{cite journal |last1=Lancini |first1=Stefano |first2=Marco |last2=Lazzari |first3=Alberto |last3=Masera |first4=Paolo |last4=Salvaneschi |title=Diagnosing Ancient Monuments with Expert Software |journal=Structural Engineering International |year=1997 |volume=7 |issue=4 |pages=288–291|url=http://dinamico2.unibg.it/lazzari/doc/structural-engineering-authors-copy.pdf|doi=10.2749/101686697780494392 |s2cid=8113173}}</ref> Mistral is a registered trade mark of [[Centro Elettrotecnico Sperimentale Italiano|CESI]].{{Missing information|section|which date ''"is still"'' in ''"Its first copy, installed in 1992 on the Dam (Italy), is still operational 24/7/365."'' refers to, all the sources cited seem to be from before 2000|date=January 2022}}

Mistral<ref name="Mistral"/> is an expert system to monitor dam safety, developed in the 1990s by Ismes in Italy. It gets data from an automatic monitoring system and performs a diagnosis of the state of the dam. Its first copy was installed in 1992 on the [[Ridracoli]] Dam in Italy. It has been installed on several other dams in Italy and abroad (e.g., [[Itaipu Dam]] in Brazil), and on landslide sites under the name of Eydenet,<ref name="Eydenet">{{cite journal |last1=Lazzari |first1=Marco |first2=Paolo |last2=Salvaneschi |title=Embedding a geographic information system in a decision support system for landslide hazard monitoring |journal=International Journal of Natural Hazards |year=1999 |volume=20 |issue=2–3 |pages=185–195 |url=http://dinamico2.unibg.it/lazzari/doc/embedding-authors-copy.pdf |doi=10.1023/A:1008187024768 |bibcode=1999NatHa..20..185L |s2cid=1746570 |archive-date=2014-10-17 |access-date=2014-10-12 |archive-url=https://web.archive.org/web/20141017145920/http://dinamico2.unibg.it/lazzari/doc/embedding-authors-copy.pdf |url-status=live }}</ref> and on monuments under the name of Kaleidos.<ref name="Kaleidos">{{cite journal |last1=Lancini |first1=Stefano |first2=Marco |last2=Lazzari |first3=Alberto |last3=Masera |first4=Paolo |last4=Salvaneschi |title=Diagnosing Ancient Monuments with Expert Software |journal=Structural Engineering International |year=1997 |volume=7 |issue=4 |pages=288–291 |url=http://dinamico2.unibg.it/lazzari/doc/structural-engineering-authors-copy.pdf |doi=10.2749/101686697780494392 |s2cid=8113173 |archive-date=2015-04-13 |access-date=2014-10-12 |archive-url=https://web.archive.org/web/20150413032053/http://dinamico2.unibg.it/lazzari/doc/structural-engineering-authors-copy.pdf |url-status=live }}</ref> Mistral is a registered trade mark of [[Centro Elettrotecnico Sperimentale Italiano|CESI]].

== See also ==

* [[AI winter]]

@@ Line 1: / Line 1: @@
-{{Short description|Computer system emulating the decision-making ability of a human expert}}
+{{Short description|Computer system emulating human expert}}
+{{For|the science fiction book series by Adrian Tchaikovsky|Expert Systems (series)}}
 [[File:Symbolics3640 Modified.JPG|200px|thumbnail|right|A [[Symbolics]] 3640 [[Lisp machine]]: an early (1984) platform for expert systems]]
 In [[artificial intelligence]] (AI), an '''expert system''' is a computer system emulating the decision-making ability of a human [[expert]].<ref name="Jackson1998">{{cite book |last1=Jackson |first1=Peter |year=1998 |title=Introduction To Expert Systems |publisher=Addison Wesley |edition=3 |isbn=978-0-201-87686-4 |page=2}}</ref>
-Expert systems are designed to solve complex problems by [[Automated reasoning system|reasoning]] through bodies of knowledge, represented mainly as [[Rule-based system|if–then rules]] rather than through conventional [[procedural programming]] code.<ref>{{cite web |url=https://www.pcmag.com/encyclopedia_term/0,2542,t=conventional+programming&i=40325,00.asp |title=Conventional programming |publisher=Pcmag.com |access-date=2013-09-15 |archive-date=2012-10-14 |archive-url=https://web.archive.org/web/20121014124656/http://www.pcmag.com/encyclopedia_term/0%2C2542%2Ct%3Dconventional+programming%26i%3D40325%2C00.asp |url-status=dead}}</ref>  Expert systems were among the first truly successful forms of AI software.<ref name="Simon & Schuster">{{cite book |last1=Russell |first1=Stuart |last2=Norvig |first2=Peter |title=Artificial Intelligence: A Modern Approach|date=1995 |publisher=Simon & Schuster |isbn=978-0-13-103805-9 |pages=22–23|url=http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf|access-date=14 June 2014 |archive-url=https://web.archive.org/web/20140505045226/http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf |archive-date=5 May 2014|url-status=dead}}</ref>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}} They were created in the 1970s and then proliferated in the 1980s,<ref name="durkinhistory">{{cite book |first= Cornelius T. |last= Leondes |title=Expert systems: the technology of knowledge management and decision making for the 21st century |year=2002 |isbn=978-0-12-443880-4 |pages=1–22}}</ref> being then widely regarded as the future of AI — before the advent of successful [[artificial neural network]]s.<ref>{{Cite news |date=2024-07-16 |title=A short history of AI |url=https://www.economist.com/schools-brief/2024/07/16/a-short-history-of-ai|access-date=2024-08-14 |newspaper=[[The Economist]]|language=en}}</ref>
+Expert systems are designed to solve complex problems by [[Automated reasoning system|reasoning]] through bodies of knowledge, represented mainly as [[Rule-based system|if–then rules]] rather than through conventional [[procedural programming]] code.<ref>{{cite web |url=https://www.pcmag.com/encyclopedia_term/0,2542,t=conventional+programming&i=40325,00.asp |title=Conventional programming |publisher=Pcmag.com |access-date=2013-09-15 |archive-date=2012-10-14 |archive-url=https://web.archive.org/web/20121014124656/http://www.pcmag.com/encyclopedia_term/0%2C2542%2Ct%3Dconventional+programming%26i%3D40325%2C00.asp |url-status=dead}}</ref>  Expert systems were among the first truly successful forms of AI software.<ref name="Simon & Schuster">{{cite book |last1=Russell |first1=Stuart |last2=Norvig |first2=Peter |title=Artificial Intelligence: A Modern Approach|date=1995 |publisher=Simon & Schuster |isbn=978-0-13-103805-9 |pages=22–23|url=http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf|access-date=14 June 2014 |archive-url=https://web.archive.org/web/20140505045226/http://stpk.cs.rtu.lv/sites/all/files/stpk/materiali/MI/Artificial%20Intelligence%20A%20Modern%20Approach.pdf |archive-date=5 May 2014|url-status=dead}}</ref>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}} They were created in the 1970s and then proliferated in the 1980s,<ref name="durkinhistory">{{cite book |first= Cornelius T. |last= Leondes |title=Expert systems: the technology of knowledge management and decision making for the 21st century |year=2002 |isbn=978-0-12-443880-4 |pages=1–22}}</ref> being then widely regarded as the future of AI — before the advent of successful [[artificial neural network]]s.<ref>{{Cite news |date=2024-07-16 |title=A short history of AI |url=https://www.economist.com/schools-brief/2024/07/16/a-short-history-of-ai |access-date=2024-08-14 |newspaper=[[The Economist]] |language=en |archive-date=2024-08-13 |archive-url=https://web.archive.org/web/20240813210432/https://www.economist.com/schools-brief/2024/07/16/a-short-history-of-ai |url-status=live }}</ref>
 An expert system is divided into two subsystems: 1) a ''[[knowledge base]]'', which represents facts and rules; and 2) an ''[[inference engine]]'', which applies the rules to the known facts to deduce new facts, and can include explaining and debugging abilities.
 == History ==
 === Early development ===
-Soon after the dawn of modern computers in the late 1940s and early 1950s, researchers started realizing the immense potential these machines had for modern society. One of the first challenges was to make such machines able to “think” like humans – in particular, making these machines able to make important decisions the way humans do. The medical–healthcare field presented the tantalizing challenge of enabling these machines to make medical diagnostic decisions.<ref name="CADsurvey">{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=A Systematic Survey of Computer-Aided Diagnosis in Medicine: Past and Present Developments |journal=Expert Systems with Applications |volume=138 |article-number=112821 |date=2019 |doi=10.1016/j.eswa.2019.112821 |s2cid=199019309}}</ref>
+Soon after the dawn of modern computers in the late 1940s and early 1950s, researchers started realizing the immense potential these machines had for modern society. One of the first challenges was to make such machines able to “think” like humans – in particular, making these machines able to make important decisions the way humans do. The medical–[[Health care|healthcare]] field presented the tantalizing challenge of enabling these machines to make medical diagnostic decisions.<ref name="CADsurvey">{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=A Systematic Survey of Computer-Aided Diagnosis in Medicine: Past and Present Developments |journal=Expert Systems with Applications |volume=138 |article-number=112821 |date=2019 |doi=10.1016/j.eswa.2019.112821 |s2cid=199019309}}</ref>
 Thus, in the late 1950s, right after the information age had fully arrived, researchers started experimenting with the prospect of using computer technology to emulate human decision making. For example, biomedical researchers started creating computer-aided systems for diagnostic applications in medicine and biology. These early diagnostic systems used patients’ symptoms and laboratory test results as inputs to generate a diagnostic outcome.<ref>{{cite journal |vauthors=Ledley RS, and Lusted LB |title=Reasoning foundations of medical diagnosis |journal=Science |date=1959 |volume=130 |issue=3366 |pages=9–21 |doi=10.1126/science.130.3366.9 |pmid=13668531 |bibcode=1959Sci...130....9L}}</ref><ref>{{cite journal |vauthors=Weiss SM, Kulikowski CA, Amarel S, Safir A |title=A model-based method for computer-aided medical decision-making |journal=Artificial Intelligence |date=1978 |volume=11 |issue=1–2 |pages=145–172 |doi=10.1016/0004-3702(78)90015-2 |citeseerx=10.1.1.464.3183}}</ref>
 These systems were often described as the early forms of expert systems. However, researchers realized that there were significant limits when using traditional methods such as flow charts,<ref>{{cite journal |vauthors=Schwartz WB |date=1970 |title=Medicine and the computer: the promise and problems of change |journal=New England Journal of Medicine |volume=283 |issue=23 |pages=1257–1264 |doi=10.1056/NEJM197012032832305 |pmid=4920342}}</ref>
-<ref>{{cite journal |vauthors=Bleich HL |date=1972 |title=Computer-based consultation: Electrolyte and acid-base disorders |journal=The American Journal of Medicine |volume=53 |issue=3 |pages=285–291 |doi=10.1016/0002-9343(72)90170-2 |pmid=4559984}}</ref> statistical pattern matching,<ref>{{cite journal |vauthors=Rosati RA, McNeer JF, Starmer CF, Mittler BS, Morris JJ, and Wallace AG |date=1975 |title=A new information system for medical practice |journal=Archives of Internal Medicine |volume=135 |issue=8 |pages=1017–1024 |doi=10.1001/archinte.1975.00330080019003 |pmid=1156062}}</ref> or probability theory.<ref>{{cite journal |vauthors=Gorry GA, Kassirer JP, Essig A, and Schwartz WB |date=1973 |title=Decision analysis as the basis for computer-aided management of acute renal failure |journal=The American Journal of Medicine |volume=55 |issue=4 |pages=473–484 |doi=10.1016/0002-9343(73)90204-0 |pmid=4582702 |s2cid=17448496}}</ref><ref>{{cite journal |vauthors=Szolovits P, Patil RS, and Schwartz WB |date=1988 |title=Artificial intelligence in medical diagnosis |journal=Annals of Internal Medicine |volume=108 |issue=1 |pages=80–87 |doi=10.7326/0003-4819-108-1-80 |pmid=3276267 |s2cid=46410202}}</ref>
+<ref>{{cite journal |vauthors=Bleich HL |date=1972 |title=Computer-based consultation: Electrolyte and acid-base disorders |journal=The American Journal of Medicine |volume=53 |issue=3 |pages=285–291 |doi=10.1016/0002-9343(72)90170-2 |pmid=4559984}}</ref> statistical [[pattern matching]],<ref>{{cite journal |vauthors=Rosati RA, McNeer JF, Starmer CF, Mittler BS, Morris JJ, and Wallace AG |date=1975 |title=A new information system for medical practice |journal=Archives of Internal Medicine |volume=135 |issue=8 |pages=1017–1024 |doi=10.1001/archinte.1975.00330080019003 |pmid=1156062}}</ref> or probability theory.<ref>{{cite journal |vauthors=Gorry GA, Kassirer JP, Essig A, and Schwartz WB |date=1973 |title=Decision analysis as the basis for computer-aided management of acute renal failure |journal=The American Journal of Medicine |volume=55 |issue=4 |pages=473–484 |doi=10.1016/0002-9343(73)90204-0 |pmid=4582702 |s2cid=17448496}}</ref><ref>{{cite journal |vauthors=Szolovits P, Patil RS, and Schwartz WB |date=1988 |title=Artificial intelligence in medical diagnosis |journal=Annals of Internal Medicine |volume=108 |issue=1 |pages=80–87 |doi=10.7326/0003-4819-108-1-80 |pmid=3276267 |s2cid=46410202}}</ref>
 === Formal introduction and later developments ===
@@ Line 18: / Line 20: @@
 This previous situation gradually led to the development of expert systems, which used knowledge-based approaches. These expert systems in medicine were the [[MYCIN]] expert system,<ref>{{cite journal |vauthors=Shortliffe EH, and Buchanan BG |title=A model of inexact reasoning in medicine |journal=Mathematical Biosciences |date=1975 |volume=23 |issue=3–4 |pages=351–379 |doi=10.1016/0025-5564(75)90047-4 |s2cid=118063112}}</ref> the [[Internist-I]] expert system<ref>{{cite journal |vauthors=Miller RA, Pople Jr HE, and Myers JD |date=1982 |title=Internist-I, an experimental computer-based diagnostic consultant for general internal medicine |journal=New England Journal of Medicine |volume=307 |issue=8 |pages=468–476 |doi=10.1056/NEJM198208193070803 |pmid=7048091}}</ref> and later, in the middle of the 1980s, the [[CADUCEUS (expert system)|CADUCEUS]].<ref>{{cite book |last1=Feigenbaum |first1=Edward |pages=1–275 |title=The fifth generation |year=1984 |publisher=Addison-Wesley |isbn=978-0451152640 |first2=Pamela |last2=McCorduck}}</ref>
-Expert systems were formally introduced around 1965 by the [[Stanford University|Stanford]] Heuristic Programming Project led by [[Edward Feigenbaum]], who is sometimes termed the "father of expert systems";<ref>{{Cite web |last=Joseph |first=Staney |date=2023-10-30 |title=The Diversity of Artificial Intelligence: How Edward Feigenbaum Developed the Expert Systems |url=https://medium.com/@staneyjoseph.in/the-diversity-of-artificial-intelligence-how-edward-feigenbaum-developed-the-expert-systems-8fe57350efb8 |access-date=2024-01-26 |website=Medium |language=en}}</ref> other key early contributors were Bruce Buchanan and Randall Davis. The Stanford researchers tried to identify domains where expertise was highly valued and complex, such as diagnosing infectious diseases ([[Mycin]]) and identifying unknown organic molecules ([[Dendral]]).<ref>{{cite book |last1=Lea |first1=Andrew S. |year=2023 |title=Digitizing Diagnosis: Medicine, Minds, and Machines in Twentieth-Century America |publisher=Johns Hopkins University Press |pages=1–256 |isbn=978-1421446813}}</ref> The idea that "intelligent systems derive their power from the knowledge they possess rather than from the specific formalisms and inference schemes they use"<ref>Edward Feigenbaum, 1977. Paraphrased by Hayes-Roth, et al.</ref> – as Feigenbaum said – was at the time a significant step forward, since the past research had been focused on heuristic computational methods, culminating in attempts to develop very general-purpose problem solvers (foremostly the conjunct work of [[Allen Newell]] and [[Herbert A. Simon|Herbert Simon]]).<ref>{{cite book |last1=Hayes-Roth |first1=Frederick |last2=Waterman |first2=Donald |last3=Lenat |first3=Douglas |author3-link=Douglas Lenat |year=1983 |pages=[https://archive.org/details/buildingexpertsy00temd/page/6 6–7] |title=Building Expert Systems |publisher=Addison-Wesley |isbn=978-0-201-10686-2 |url=https://archive.org/details/buildingexpertsy00temd/page/6}}</ref> Expert systems became some of the first truly successful forms of [[artificial intelligence]] (AI) software.<ref name="Simon & Schuster"/>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}}
+Expert systems were formally introduced around 1965 by the [[Stanford University|Stanford]] Heuristic Programming Project led by [[Edward Feigenbaum]], who is sometimes termed the "father of expert systems";<ref>{{Cite web |last=Joseph |first=Staney |date=2023-10-30 |title=The Diversity of Artificial Intelligence: How Edward Feigenbaum Developed the Expert Systems |url=https://medium.com/@staneyjoseph.in/the-diversity-of-artificial-intelligence-how-edward-feigenbaum-developed-the-expert-systems-8fe57350efb8 |access-date=2024-01-26 |website=Medium |language=en |archive-date=2024-01-26 |archive-url=https://web.archive.org/web/20240126214030/https://medium.com/@staneyjoseph.in/the-diversity-of-artificial-intelligence-how-edward-feigenbaum-developed-the-expert-systems-8fe57350efb8 |url-status=live }}</ref> other key early contributors were Bruce Buchanan and Randall Davis. The Stanford researchers tried to identify domains where expertise was highly valued and complex, such as diagnosing infectious diseases ([[Mycin]]) and identifying unknown organic molecules ([[Dendral]]).<ref>{{cite book |last1=Lea |first1=Andrew S. |year=2023 |title=Digitizing Diagnosis: Medicine, Minds, and Machines in Twentieth-Century America |publisher=Johns Hopkins University Press |pages=1–256 |isbn=978-1421446813}}</ref> The idea that "intelligent systems derive their power from the knowledge they possess rather than from the specific formalisms and inference schemes they use"<ref>Edward Feigenbaum, 1977. Paraphrased by Hayes-Roth, et al.</ref> – as Feigenbaum said – was at the time a significant step forward, since the past research had been focused on heuristic computational methods, culminating in attempts to develop very general-purpose problem solvers (foremostly the conjunct work of [[Allen Newell]] and [[Herbert A. Simon|Herbert Simon]]).<ref>{{cite book |last1=Hayes-Roth |first1=Frederick |last2=Waterman |first2=Donald |last3=Lenat |first3=Douglas |author3-link=Douglas Lenat |year=1983 |pages=[https://archive.org/details/buildingexpertsy00temd/page/6 6–7] |title=Building Expert Systems |publisher=Addison-Wesley |isbn=978-0-201-10686-2 |url=https://archive.org/details/buildingexpertsy00temd/page/6}}</ref> Expert systems became some of the first truly successful forms of [[artificial intelligence]] (AI) software.<ref name="Simon & Schuster"/>{{sfn|Luger|Stubblefield|2004|pp=227–331}}{{sfn|Nilsson|1998|loc=chpt. 17.4}}{{sfn|McCorduck|2004|pp=327–335, 434–435}}{{sfn|Crevier|1993|pp=145–62, 197−203}}
 Research on expert systems was also active in Europe. In the US, the focus tended to be on the use of [[Production system (computer science)|production rule systems]], first on systems hard coded on top of [[Lisp (programming language)|Lisp]] programming environments and then on expert system shells developed by vendors such as [[IntelliCorp (software)|Intellicorp]]. In Europe, research focused more on systems and expert systems shells developed in [[Prolog]]. The advantage of Prolog systems was that they employed a form of [[rule-based system|rule-based programming]] that was based on [[logic programming|formal logic]].<ref>George F. Luger and William A. Stubblefield, Benjamin/Cummings Publishers, Rule-Based Expert System Shell: example of code using the Prolog rule-based expert system shell</ref><ref>[http://promethee.philo.ulg.ac.be/engdep1/download/prolog/htm_docs/prolog.htm A. Michiels] {{Webarchive|url=https://web.archive.org/web/20120402132354/http://promethee.philo.ulg.ac.be/engdep1/download/prolog/htm_docs/prolog.htm |date=2012-04-02}}, Université de Liège, Belgique: "PROLOG, the first declarative language</ref>
-One such early expert system shell based on Prolog was APES.<ref name="APES">{{citation |url=https://www.ojp.gov/ncjrs/virtual-library/abstracts/investigating-apes-augmented-prolog-expert-system |title=Investigating with APES (Augmented Prolog Expert System) |access-date=2024-01-03}}</ref>
+One such early expert system shell based on [[Prolog]] was APES.<ref name="APES">{{citation |url=https://www.ojp.gov/ncjrs/virtual-library/abstracts/investigating-apes-augmented-prolog-expert-system |title=Investigating with APES (Augmented Prolog Expert System) |access-date=2024-01-03 |archive-date=2024-01-03 |archive-url=https://web.archive.org/web/20240103115654/https://www.ojp.gov/ncjrs/virtual-library/abstracts/investigating-apes-augmented-prolog-expert-system |url-status=live }}</ref>
-One of the first use cases of [[Prolog]] and APES was in the legal area namely, the encoding of a large portion of the British Nationality Act. Lance Elliot wrote: "The British Nationality Act was passed in 1981 and shortly thereafter was used as a means of showcasing the efficacy of using Artificial Intelligence (AI) techniques and technologies, doing so to explore how the at-the-time newly enacted statutory law might be encoded into a computerized logic-based formalization. A now oft-cited research paper entitled “The British Nationality Act as a Logic Program” was published in 1986 and subsequently became a hallmark for subsequent work in AI and the law."<ref name="AI & Law">
+One of the first use cases of Prolog and APES was in the legal area namely, the encoding of a large portion of the British Nationality Act. Lance Elliot wrote: "[[British Nationality Act 1981|The British Nationality Act]] was passed in 1981 and shortly thereafter was used as a means of showcasing the efficacy of using Artificial Intelligence (AI) techniques and technologies, doing so to explore how the at-the-time newly enacted statutory law might be encoded into a computerized logic-based formalization. A now oft-cited research paper entitled “The British Nationality Act as a Logic Program” was published in 1986 and subsequently became a hallmark for subsequent work in AI and the law."<ref name="AI & Law">
-{{citation |url=https://lance-eliot.medium.com/ai-law-british-nationality-act-unexpectedly-spurred-ai-and-law-404aea03386a |title=AI & Law: British Nationality Act Unexpectedly Spurred AI And Law |date=17 April 2021 |access-date=2023-11-13}}</ref><ref name="BNA">{{cite journal |doi=10.1145/5689.5920 |author=M.J. Sergot and F. Sadri and R.A. Kowalski and F. Kriwaczek and P. Hammond and H.T. Cory |date=May 1986 |title=The British Nationality Act as a Logic Program |journal=Communications of the ACM |volume=29 |number=5 |pages=370–386}}</ref>
+{{citation |last1=Eliot |first1=Lance |url=https://lance-eliot.medium.com/ai-law-british-nationality-act-unexpectedly-spurred-ai-and-law-404aea03386a |title=AI & Law: British Nationality Act Unexpectedly Spurred AI And Law |work=Medium |date=17 April 2021 |access-date=2023-11-13}}</ref><ref name="BNA">{{cite journal |doi=10.1145/5689.5920 |author=M.J. Sergot and F. Sadri and R.A. Kowalski and F. Kriwaczek and P. Hammond and H.T. Cory |date=May 1986 |title=The British Nationality Act as a Logic Program |journal=Communications of the ACM |volume=29 |number=5 |pages=370–386}}</ref>
 In the 1980s, expert systems proliferated. Universities offered expert system courses and two-thirds of the [[Fortune 500]] companies applied the technology in daily business activities.<ref name=durkinhistory/><ref>Durkin, J. Expert Systems: Catalog of Applications. Intelligent Computer Systems, Inc., Akron, OH, 1993.</ref> Interest was international with the [[Fifth generation computer|Fifth Generation Computer Systems project]] in Japan and increased research funding in Europe.
-In 1981, the first [[IBM PC]], with the [[IBM PC DOS|PC DOS]] operating system, was introduced.<ref>{{Cite web |title=The IBM PC - CHM Revolution |url=https://www.computerhistory.org/revolution/personal-computers/17/301 |access-date=2024-01-26 |website=www.computerhistory.org}}</ref> The imbalance between the low cost of the relatively powerful chips in the PC, compared to the much more expensive cost of processing power in the mainframes that dominated the corporate IT world at the time, created a new type of architecture for corporate computing, termed the [[client–server model]].<ref>{{cite book |last1=Orfali |first1=Robert |year=1996 |title=The Essential Client/Server Survival Guide |publisher=Wiley Computer Publishing |location=New York |isbn=978-0-471-15325-2 |pages=[https://archive.org/details/essentialclients00orfa/page/1 1–10] |url=https://archive.org/details/essentialclients00orfa/page/1}}</ref> Calculations and reasoning could be performed at a fraction of the price of a mainframe using a PC. This model also enabled business units to bypass corporate IT departments and directly build their own applications. As a result, client-server had a tremendous impact on the expert systems market. Expert systems were already outliers in much of the business world, requiring new skills that many IT departments did not have and were not eager to develop. They were a natural fit for new PC-based shells that promised to put application development into the hands of end users and experts. Until then, the main development environment for expert systems had been high end [[Lisp machine]]s from [[Xerox]], [[Symbolics]], and [[Texas Instruments]]. With the rise of the PC and client-server computing, vendors such as Intellicorp and Inference Corporation shifted their priorities to developing PC-based tools. Also, new vendors, often financed by [[venture capital]] (such as Aion Corporation, [[Neuron Data]], Exsys, [[VP-Expert]], and many others<ref>{{cite book |last=Hurwitz |first=Judith |title=Smart or Lucky: How Technology Leaders Turn Chance into Success |year=2011 |publisher=John Wiley & Son |isbn=978-1118033784 |page=164|url=https://books.google.com/books?id=3KrTQzQHl7AC&q=expert+systems+failed+to+live+up+to+hype&pg=PA164|access-date=29 November 2013}}</ref><ref>{{cite journal |first=Robert J. |last=Dunn |date=September 30, 1985 |title=Expandable Expertise for Everyday Users |page=30 |journal=InfoWorld |volume=7 |issue=39 |url=https://books.google.com/books?id=iS8EAAAAMBAJ&pg=PA30 |access-date=2011-03-13}}</ref>), started appearing regularly.
+In 1981, the first [[IBM PC]], with the [[IBM PC DOS|PC DOS]] operating system, was introduced.<ref>{{Cite web |title=The IBM PC - CHM Revolution |url=https://www.computerhistory.org/revolution/personal-computers/17/301 |access-date=2024-01-26 |website=www.computerhistory.org |archive-date=2024-01-26 |archive-url=https://web.archive.org/web/20240126214029/https://www.computerhistory.org/revolution/personal-computers/17/301 |url-status=live }}</ref> The imbalance between the low cost of the relatively powerful chips in the PC, compared to the much more expensive cost of processing power in the mainframes that dominated the corporate IT world at the time, created a new type of architecture for corporate computing, termed the [[client–server model]].<ref>{{cite book |last1=Orfali |first1=Robert |year=1996 |title=The Essential Client/Server Survival Guide |publisher=Wiley Computer Publishing |location=New York |isbn=978-0-471-15325-2 |pages=[https://archive.org/details/essentialclients00orfa/page/1 1–10] |url=https://archive.org/details/essentialclients00orfa/page/1}}</ref> Calculations and reasoning could be performed at a fraction of the price of a mainframe using a PC. This model also enabled business units to bypass corporate IT departments and directly build their own applications. As a result, client-server had a tremendous impact on the expert systems market. Expert systems were already outliers in much of the business world, requiring new skills that many IT departments did not have and were not eager to develop. They were a natural fit for new PC-based shells that promised to put application development into the hands of end users and experts. Until then, the main development environment for expert systems had been high end [[Lisp machine]]s from [[Xerox]], [[Symbolics]], and [[Texas Instruments]]. With the rise of the PC and client-server computing, vendors such as Intellicorp and Inference Corporation shifted their priorities to developing PC-based tools. Also, new vendors, often financed by [[venture capital]]<ref>{{cite book |last=Hurwitz |first=Judith |title=Smart or Lucky: How Technology Leaders Turn Chance into Success |year=2011 |publisher=John Wiley & Son |isbn=978-1118033784 |page=164|url=https://books.google.com/books?id=3KrTQzQHl7AC&q=expert+systems+failed+to+live+up+to+hype&pg=PA164|access-date=29 November 2013}}</ref><ref>{{cite journal |first=Robert J. |last=Dunn |date=September 30, 1985 |title=Expandable Expertise for Everyday Users |page=30 |journal=InfoWorld |volume=7 |issue=39 |url=https://books.google.com/books?id=iS8EAAAAMBAJ&pg=PA30 |access-date=2011-03-13}}</ref> started appearing regularly.
 The first expert system to be used in a design capacity for a large-scale product was the Synthesis of Integral Design (SID) software program, developed in 1982. Written in [[Lisp (programming language)|Lisp]], SID generated 93% of the [[VAX 9000]] CPU logic gates.<ref name="SIDDTJ">{{cite journal |last1=Gibson |first1=Carl S. |display-authors=etal |title=VAX 9000 Series |journal=Digital Technical Journal of Digital Equipment Corporation |volume=2 |issue=4, Fall 1990 |pages=118–129}}</ref> Input to the software was a set of rules created by several expert logic designers. SID expanded the rules and generated software [[logic synthesis]] routines many times the size of the rules themselves. Surprisingly, the combination of these rules resulted in an overall design that exceeded the capabilities of the experts themselves, and in many cases out-performed the human counterparts. While some rules contradicted others, top-level control parameters for speed and area provided the tie-breaker. The program was highly controversial but used nevertheless due to project budget constraints. It was terminated by logic designers after the VAX 9000 project completion.
@@ Line 34: / Line 36: @@
 During the years before the middle of the 1970s, the expectations of what expert systems can accomplish in many fields tended to be extremely optimistic. At the start of these early studies, researchers were hoping to develop entirely automatic (i.e., completely computerized) expert systems. The expectations of people of what computers can do were frequently too idealistic. This situation radically changed after [[Richard M. Karp]] published his breakthrough paper: “Reducibility among Combinatorial Problems” in the early 1970s.<ref>{{cite book |last1=Karp |first1=Richard M. |year=1972 |chapter=Reducibility Among Combinatorial Problems |chapter-url=http://www.cs.berkeley.edu/~luca/cs172/karp.pdf |title=Complexity of Computer Computations |editor1-last=Miller |editor1-first=R. E. |editor2-last=Thatcher |editor2-first=J. W. |publisher=New York: Plenum |pages=85–103 |access-date=2020-01-24 |archive-date=2011-06-29 |url-status=dead |archive-url=https://web.archive.org/web/20110629023717/http://www.cs.berkeley.edu/~luca/cs172/karp.pdf}}</ref> Thanks to Karp's work, together with other scholars, like Hubert L. Dreyfus,<ref>{{cite book |author=Hubert L. Dreyfus |title=What Computers Still Can't Do |publisher=Cambridge, Massachusetts: The MIT Press |year=1972}}</ref> it became clear that there are certain limits and possibilities when one designs computer algorithms. His findings describe what computers can do and what they cannot do. Many of the computational problems related to this type of expert systems have certain pragmatic limits. These findings laid down the groundwork that led to the next developments in the field.<ref name="CADsurvey"/>
-In the 1990s and beyond, the term ''expert system'' and the idea of a standalone AI system mostly dropped from the IT lexicon. There are two interpretations of this. One is that "expert systems failed": the IT world moved on because expert systems did not deliver on their over hyped promise.<ref>{{Cite web |url=http://www.ainewsletter.com/newsletters/aix_0501.htm#w |title=AI Expert Newsletter: W is for Winter |access-date=2013-11-29 |archive-url=https://web.archive.org/web/20131109201636/http://www.ainewsletter.com/newsletters/aix_0501.htm#w |archive-date=2013-11-09 |url-status=dead}}</ref><ref>{{cite journal |last1=Leith |first1=P. |date=2010 |url=http://ejlt.org//article/view/14/1 |title=The rise and fall of the legal expert system |journal=European Journal of Law and Technology |volume=1 |issue=1 |access-date=2020-01-24 |url-status=dead |archive-date=2016-03-04 |archive-url=https://web.archive.org/web/20160304124042/http://ejlt.org//article/view/14/1}}</ref> The other is the mirror opposite, that expert systems were simply victims of their success: as IT professionals grasped concepts such as rule engines, such tools migrated from being standalone tools for developing special purpose ''expert'' systems, to being one of many standard tools.<ref>{{cite journal |last1=Haskin |first1=David |date=January 16, 2003 |title=Years After Hype, 'Expert Systems' Paying Off for Some |journal=Datamation |url=http://www.datamation.com/netsys/article.php/1570851/Years-After-Hype-Expert-Systems-Paying-Off-For-Some.htm|access-date=29 November 2013}}</ref> Other researchers suggest that Expert Systems caused inter-company power struggles when the IT organization lost its exclusivity in software modifications to users or Knowledge Engineers.<ref>{{cite journal |last1=Romem |first1=Yoram |title=The Social Construction of Expert Systems |journal=Human Systems Management |date=2007 |volume=26 |issue=4 |pages=291–309 |doi=10.3233/HSM-2007-26406 |url=https://www.researchgate.net/publication/228630533}}</ref>
+In the 1990s and beyond, the term ''expert system'' and the idea of a standalone AI system mostly dropped from the IT lexicon. There are two interpretations of this. One is that "expert systems failed": the IT world moved on because expert systems did not deliver on their over hyped promise.<ref>{{Cite web |url=http://www.ainewsletter.com/newsletters/aix_0501.htm#w |title=AI Expert Newsletter: W is for Winter |access-date=2013-11-29 |archive-url=https://web.archive.org/web/20131109201636/http://www.ainewsletter.com/newsletters/aix_0501.htm#w |archive-date=2013-11-09 |url-status=dead}}</ref><ref>{{cite journal |last1=Leith |first1=P. |date=2010 |url=http://ejlt.org//article/view/14/1 |title=The rise and fall of the legal expert system |journal=European Journal of Law and Technology |volume=1 |issue=1 |access-date=2020-01-24 |url-status=dead |archive-date=2016-03-04 |archive-url=https://web.archive.org/web/20160304124042/http://ejlt.org//article/view/14/1}}</ref> The other is the mirror opposite, that expert systems were simply victims of their success: as IT professionals grasped concepts such as rule engines, such tools migrated from being standalone tools for developing special purpose ''expert'' systems, to being one of many standard tools.<ref>{{cite journal |last1=Haskin |first1=David |date=January 16, 2003 |title=Years After Hype, 'Expert Systems' Paying Off for Some |journal=Datamation |url=http://www.datamation.com/netsys/article.php/1570851/Years-After-Hype-Expert-Systems-Paying-Off-For-Some.htm |access-date=29 November 2013 |archive-date=3 December 2013 |archive-url=https://web.archive.org/web/20131203013322/http://www.datamation.com/netsys/article.php/1570851/Years-After-Hype-Expert-Systems-Paying-Off-For-Some.htm |url-status=live }}</ref> Other researchers suggest that Expert Systems caused inter-company power struggles when the IT organization lost its exclusivity in software modifications to users or Knowledge Engineers.<ref>{{cite journal |last1=Romem |first1=Yoram |title=The Social Construction of Expert Systems |journal=Human Systems Management |date=2007 |volume=26 |issue=4 |pages=291–309 |doi=10.3233/HSM-2007-26406 |url=https://www.researchgate.net/publication/228630533}}</ref>
-In the first decade of the 2000s, there was a "resurrection" for the technology, while using the term ''[[rule-based system]]s'', with significant success stories and adoption.<ref>{{cite news |last1=Voelker |first1=Michael P. |date=October 18, 2005 |url=https://www.informationweek.com/information-management/business-makes-the-rules |title=Business Makes the Rules |work=Information Week}}</ref> Many of the leading major business application suite vendors (such as [[SAP (software)|SAP]], [[Siebel Systems|Siebel]], and [[Oracle Corporation|Oracle]]) integrated expert system abilities into their suite of products as a way to specify business logic. Rule engines are no longer simply for defining the rules an expert would use but for any type of complex, volatile, and critical business logic; they often go hand in hand with business process automation and integration environments.<ref>{{cite web |last=SAP News Desk |title=SAP News Desk IntelliCorp Announces Participation in SAP EcoHub |url=http://laszlo.sys-con.com/node/946452 |work=laszlo.sys-con.com |publisher=LaszloTrack |access-date=29 November 2013 |archive-date=3 December 2013 |archive-url=https://web.archive.org/web/20131203002523/http://laszlo.sys-con.com/node/946452 |url-status=dead}}</ref><ref>{{cite web |last=Pegasystems |title=Smart BPM Requires Smart Business Rules |url=http://www.pega.com/business-rules |work=pega.com |access-date=29 November 2013}}</ref><ref>{{cite book |last1=Zhao |first1=Kai |first2=Shi |last2=Ying |first3=Linlin |last3=Zhang |first4=Luokai |last4=Hu |title=2010 International Conference on Future Information Technology and Management Engineering |chapter=Achieving business process and business rules integration using SPL |date=9–10 Oct 2010 |volume=2 |pages=329–332  |publisher=IEEE |doi=10.1109/fitme.2010.5656297 |isbn=978-1-4244-9087-5}}</ref>
+In the first decade of the 2000s, there was a "resurrection" for the technology, while using the term ''[[rule-based system]]s'', with significant success stories and adoption.<ref>{{cite news |last1=Voelker |first1=Michael P. |date=October 18, 2005 |url=https://www.informationweek.com/information-management/business-makes-the-rules |title=Business Makes the Rules |work=Information Week |archive-date=December 23, 2022 |access-date=December 23, 2022 |archive-url=https://web.archive.org/web/20221223174411/https://www.informationweek.com/information-management/business-makes-the-rules |url-status=live }}</ref> Many of the leading major business application suite vendors (such as [[SAP (software)|SAP]], [[Siebel Systems|Siebel]], and [[Oracle Corporation|Oracle]]) integrated expert system abilities into their suite of products as a way to specify business logic. Rule engines are no longer simply for defining the rules an expert would use but for any type of complex, volatile, and critical business logic; they often go hand in hand with business process automation and integration environments.<ref>{{cite web |last=SAP News Desk |title=SAP News Desk IntelliCorp Announces Participation in SAP EcoHub |url=http://laszlo.sys-con.com/node/946452 |work=laszlo.sys-con.com |publisher=LaszloTrack |access-date=29 November 2013 |archive-date=3 December 2013 |archive-url=https://web.archive.org/web/20131203002523/http://laszlo.sys-con.com/node/946452 |url-status=dead}}</ref><ref>{{cite web |last=Pegasystems |title=Smart BPM Requires Smart Business Rules |url=http://www.pega.com/business-rules |work=pega.com |access-date=29 November 2013 |archive-date=9 November 2013 |archive-url=https://web.archive.org/web/20131109110005/http://www.pega.com/business-rules |url-status=live }}</ref><ref>{{cite book |last1=Zhao |first1=Kai |first2=Shi |last2=Ying |first3=Linlin |last3=Zhang |first4=Luokai |last4=Hu |title=2010 International Conference on Future Information Technology and Management Engineering |chapter=Achieving business process and business rules integration using SPL |date=9–10 Oct 2010 |volume=2 |pages=329–332  |publisher=IEEE |doi=10.1109/fitme.2010.5656297 |isbn=978-1-4244-9087-5}}</ref>
 === Current approaches to expert systems ===
-The limits of prior type of expert systems prompted researchers to develop new types of approaches. They have developed more efficient, flexible, and powerful methods to simulate the human decision-making process. Some of the approaches that researchers have developed are based on new methods of artificial intelligence (AI), and in particular in [[machine learning]] and [[data mining]] approaches with a feedback mechanism.<ref>{{Cite journal |last1=Chung |first1=Junyoung |last2=Gulcehre |first2=Caglar |last3=Cho |first3=Kyunghyun |last4=Bengio |first4=Yoshua |date=2015-06-01 |title=Gated Feedback Recurrent Neural Networks |url=http://proceedings.mlr.press/v37/chung15.html |journal=International Conference on Machine Learning |language=en |publisher=PMLR |pages=2067–2075 |arxiv=1502.02367}}</ref>{{failed verification|reason=The article doesn't mention expert systems at all.|date=May 2024}} [[Recurrent neural network]]s often take advantage of such mechanisms. Related is the discussion on the disadvantages section.
+The limits of prior types of expert systems prompted researchers to develop new types of approaches. They have developed more efficient, flexible, and powerful methods to simulate the human decision-making process. Some of the approaches that researchers have developed are based on new methods of artificial intelligence (AI), and in particular in [[machine learning]] and [[data mining]] approaches with a feedback mechanism.<ref>{{Cite journal |last1=Chung |first1=Junyoung |last2=Gulcehre |first2=Caglar |last3=Cho |first3=Kyunghyun |last4=Bengio |first4=Yoshua |date=2015-06-01 |title=Gated Feedback Recurrent Neural Networks |url=http://proceedings.mlr.press/v37/chung15.html |journal=International Conference on Machine Learning |language=en |publisher=PMLR |pages=2067–2075 |arxiv=1502.02367 |archive-date=2021-05-24 |access-date=2021-05-24 |archive-url=https://web.archive.org/web/20210524143259/http://proceedings.mlr.press/v37/chung15.html |url-status=live }}</ref>{{failed verification|reason=The article doesn't mention expert systems at all.|date=May 2024}} [[Recurrent neural network]]s often take advantage of such mechanisms. Related is the discussion on the disadvantages section.
 Modern systems can incorporate new knowledge more easily and thus update themselves easily. Such systems can generalize from existing knowledge better and deal with vast amounts of complex data. Related is the subject of [[big data]] here. Sometimes these types of expert systems are called "intelligent systems."<ref name="CADsurvey"/>
@@ Line 46: / Line 48: @@
 == Software architecture ==
-[[File:BackwardChaining David C England 1990 p21.gif|thumb|450px|Illustrating example of ''backward chaining'' from a 1990 Master's Thesis<ref>{{cite thesis |type=MSc |url=https://commons.wikimedia.org/wiki/File:An_expert_system_for_the_management_of_hazardous_materials_at_a_Naval_Supply_Center_(IA_anexpertsystemfo1094530640).pdf |last1=England |first1=David C. |date=June 1990 |title=An Expert System for the Management of Hazardous Materials at a Naval Supply Center |publisher=Naval Postgraduate School Monterey/CA |page=21}}</ref>]]
+[[File:BackwardChaining David C England 1990 p21.gif|thumb|450px|Illustrating example of ''backward chaining'' from a 1990 Master's Thesis<ref>{{cite thesis |type=MSc |url=https://commons.wikimedia.org/wiki/File:An_expert_system_for_the_management_of_hazardous_materials_at_a_Naval_Supply_Center_(IA_anexpertsystemfo1094530640).pdf |last1=England |first1=David C. |date=June 1990 |title=An Expert System for the Management of Hazardous Materials at a Naval Supply Center |publisher=Naval Postgraduate School Monterey/CA |page=21 |archive-date=2021-10-18 |access-date=2020-09-01 |archive-url=https://web.archive.org/web/20211018080757/https://commons.wikimedia.org/wiki/File:An_expert_system_for_the_management_of_hazardous_materials_at_a_Naval_Supply_Center_(IA_anexpertsystemfo1094530640).pdf |url-status=live }}</ref>]]
-An expert system is an example of a [[knowledge-based system]]. Expert systems were the first commercial systems to use a knowledge-based architecture. In general view, an expert system includes the following components: a [[knowledge base]], an [[inference engine]], an explanation facility, a knowledge acquisition facility, and a user interface.<ref name=":0">{{Cite journal |last=Kiryanov |first=Denis Aleksandrovich|date=2021-12-21 |title=Hybrid categorical expert system for use in content aggregation |journal=Software Systems and Computational Methods |issue=4 |pages=1–22|doi=10.7256/2454-0714.2021.4.37019 |s2cid=245498498 |issn=2454-0714|doi-access=free}}</ref><ref>{{cite web |last=Smith |first=Reid|date=May 8, 1985 |title=Knowledge-Based Systems Concepts, Techniques, Examples|url=http://www.reidgsmith.com/Knowledge-Based_Systems_-_Concepts_Techniques_Examples_08-May-1985.pdf|access-date=9 November 2013|website=Reid G. Smith}}</ref>
+An expert system is an example of a [[knowledge-based system]]. Expert systems were the first commercial systems to use a knowledge-based architecture. In general view, an expert system includes the following components: a [[knowledge base]], an [[inference engine]], an explanation facility, a knowledge acquisition facility, and a user interface.<ref name=":0">{{Cite journal |last=Kiryanov |first=Denis Aleksandrovich|date=2021-12-21 |title=Hybrid categorical expert system for use in content aggregation |journal=Software Systems and Computational Methods |issue=4 |pages=1–22|doi=10.7256/2454-0714.2021.4.37019 |s2cid=245498498 |issn=2454-0714|doi-access=free}}</ref><ref>{{cite web|last=Smith|first=Reid|date=May 8, 1985|title=Knowledge-Based Systems Concepts, Techniques, Examples|url=http://www.reidgsmith.com/Knowledge-Based_Systems_-_Concepts_Techniques_Examples_08-May-1985.pdf|access-date=9 November 2013|website=Reid G. Smith|archive-date=10 February 2023|archive-url=https://web.archive.org/web/20230210113921/http://www.reidgsmith.com/Knowledge-Based_Systems_-_Concepts_Techniques_Examples_08-May-1985.pdf|url-status=live}}</ref>
-The knowledge base represents facts about the world. In early expert systems such as Mycin and Dendral, these facts were represented mainly as flat assertions about variables. In later expert systems developed with commercial shells, the knowledge base took on more structure and used concepts from [[object-oriented programming]]. The world was represented as [[Class (computer programming)|classes, subclasses]], and [[Instance (computer science)|instances]] and [[Assertion (software development)|assertions]] were replaced by values of object instances. The rules worked by querying and asserting values of the objects.
+The knowledge base represents facts about the world. In early expert systems such as Mycin and Dendral, these facts were represented mainly as flat assertions about variables. In later expert systems developed with commercial shells, the knowledge base took on more structure and used concepts from [[object-oriented programming]]. The world was represented as [[Class (programming)|classes, subclasses]], and [[Instance (computer science)|instances]] and [[Assertion (software development)|assertions]] were replaced by values of object instances. The rules worked by querying and asserting values of the objects.
 The inference engine is an [[automated reasoning system]] that evaluates the current state of the knowledge-base, applies relevant rules, and then asserts new knowledge into the knowledge base. The inference engine may also include abilities for explanation, so that it can explain to a user the chain of reasoning used to arrive at a particular conclusion by tracing back over the firing of rules that resulted in the assertion.<ref name="Hayes-Roth 1983">{{cite book |last1=Hayes-Roth |first1=Frederick |last2=Waterman |first2=Donald |last3=Lenat |first3=Douglas |author3-link=Douglas Lenat |year=1983 |title=Building Expert Systems |publisher=Addison-Wesley |isbn=978-0-201-10686-2 |url=https://archive.org/details/buildingexpertsy00temd}}</ref>
@@ Line 61: / Line 63: @@
 Backward chaining is a bit less straight forward. In backward chaining the system looks at possible conclusions and works backward to see if they might be true. So if the system was trying to determine if Mortal(Socrates) is true it would find R1 and query the knowledge base to see if Man(Socrates) is true. One of the early innovations of expert systems shells was to integrate inference engines with a user interface. This could be especially powerful with backward chaining. If the system needs to know a particular fact but does not, then it can simply generate an input screen and ask the user if the information is known. So in this example, it could use R1 to ask the user if Socrates was a Man and then use that new information accordingly.
-The use of rules to explicitly represent knowledge also enabled explanation abilities. In the simple example above if the system had used R1 to assert that Socrates was Mortal and a user wished to understand why Socrates was mortal they could query the system and the system would look back at the rules which fired to cause the assertion and present those rules to the user as an explanation. In English, if the user asked "Why is Socrates Mortal?" the system would reply "Because all men are mortal and Socrates is a man". A significant area for research was the generation of explanations from the knowledge base in natural English rather than simply by showing the more formal but less intuitive rules.<ref>[http://www.ccis2k.org/iajit/PDF/vol.4,no.1/9-Nabil.pdf Nabil Arman], Polytechnic University of Palestine, January 2007, Fault Detection in Dynamic Rule Bases Using Spanning Trees and Disjoin Sets: ""</ref>
+The use of rules to explicitly represent knowledge also enabled explanation abilities. In the simple example above if the system had used R1 to assert that Socrates was Mortal and a user wished to understand why Socrates was mortal they could query the system and the system would look back at the rules which fired to cause the assertion and present those rules to the user as an explanation. In English, if the user asked "Why is Socrates Mortal?" the system would reply "Because all men are mortal and Socrates is a man". A significant area for research was the generation of explanations from the knowledge base in natural English rather than simply by showing the more formal but less intuitive rules.<ref>[http://www.ccis2k.org/iajit/PDF/vol.4,no.1/9-Nabil.pdf Nabil Arman] {{Webarchive|url=https://web.archive.org/web/20120425141837/http://www.ccis2k.org/iajit/PDF/vol.4,no.1/9-Nabil.pdf |date=2012-04-25 }}, Polytechnic University of Palestine, January 2007, Fault Detection in Dynamic Rule Bases Using Spanning Trees and Disjoin Sets: ""</ref>
 As expert systems evolved, many new techniques were incorporated into various types of inference engines.<ref>{{cite journal |last=Mettrey |first=William |title=An Assessment of Tools for Building Large Knowledge-Based Systems |journal=AI Magazine |year=1987 |volume=8 |issue=4|url=http://www.aaai.org/ojs/index.php/aimagazine/article/viewArticle/625|access-date=2013-11-29 |archive-url=https://web.archive.org/web/20131110022104/http://www.aaai.org/ojs/index.php/aimagazine/article/viewArticle/625 |archive-date=2013-11-10|url-status=dead}}</ref> Some of the most important of these were:
@@ Line 67: / Line 69: @@
 * Hypothetical reasoning. In this, the knowledge base can be divided up into many possible views, a.k.a. worlds. This allows the inference engine to explore multiple possibilities in parallel. For example, the system may want to explore the consequences of both assertions, what will be true if Socrates is a Man and what will be true if he is not?
 * Uncertainty systems. One of the first extensions of simply using rules to represent knowledge was also to associate a probability with each rule. So, not to assert that Socrates is mortal, but to assert Socrates ''may'' be mortal with some probability value. Simple probabilities were extended in some systems with sophisticated mechanisms for uncertain reasoning, such as [[fuzzy logic]], and combination of probabilities.
-* [[Ontology (information science)|Ontology]] classification. With the addition of object classes to the knowledge base, a new type of reasoning was possible. Along with reasoning simply about object values, the system could also reason about object structures. In this simple example, Man can represent an object class and R1 can be redefined as a rule that defines the class of all men. These types of special purpose inference engines are termed [[Deductive classifier|classifiers]]. Although they were not highly used in expert systems, classifiers are very powerful for unstructured volatile domains, and are a key technology for the Internet and the emerging [[Semantic Web]].<ref>{{cite journal |last1=MacGregor |first1=Robert |date=June 1991 |title=Using a description classifier to enhance knowledge representation |journal=IEEE Expert |volume=6 |issue=3 |doi=10.1109/64.87683 |pages=41–46 |s2cid=29575443}}</ref><ref>{{cite journal |last1=Berners-Lee |first1=Tim |last2=Hendler |first2=James |last3=Lassila |first3=Ora |date=May 17, 2001 |title=The Semantic Web A new form of Web content that is meaningful to computers will unleash a revolution of new possibilities |journal=Scientific American |url=http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |doi=10.1038/scientificamerican0501-34 |volume=284 |issue=5 |pages=34–43 |url-status=dead |archive-url=https://web.archive.org/web/20130424071228/http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |archive-date=April 24, 2013 |url-access=subscription}}</ref>
+* [[Ontology (information science)|Ontology]] classification. With the addition of object classes to the knowledge base, a new type of reasoning was possible. Along with reasoning simply about object values, the system could also reason about object structures. In this simple example, Man can represent an object class and R1 can be redefined as a rule that defines the class of all men. These types of special purpose inference engines are termed [[Deductive classifier|classifiers]]. Although they were not highly used in expert systems, classifiers are very powerful for unstructured volatile domains, and are a key technology for the Internet and the emerging [[Semantic Web]].<ref>{{cite journal |last1=MacGregor |first1=Robert |date=June 1991 |title=Using a description classifier to enhance knowledge representation |journal=IEEE Expert |volume=6 |issue=3 |doi=10.1109/64.87683 |pages=41–46 |bibcode=1991IExp....6...41M |s2cid=29575443}}</ref><ref>{{cite journal |last1=Berners-Lee |first1=Tim |last2=Hendler |first2=James |last3=Lassila |first3=Ora |date=May 17, 2001 |title=The Semantic Web A new form of Web content that is meaningful to computers will unleash a revolution of new possibilities |journal=Scientific American |url=http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |doi=10.1038/scientificamerican0501-34 |volume=284 |issue=5 |pages=34–43 |url-status=dead |archive-url=https://web.archive.org/web/20130424071228/http://www.cs.umd.edu/~golbeck/LBSC690/SemanticWeb.html |archive-date=April 24, 2013 |url-access=subscription}}</ref>
 == Advantages ==
@@ Line 90: / Line 92: @@
 Another major challenge of expert systems emerges when the size of the knowledge base increases. This causes the processing complexity to increase. For instance, when an expert system with 100 million rules was envisioned as the ultimate expert system, it became obvious that such system would be too complex and it would face too many computational problems.<ref>{{cite book |last1=Lenat |first1=Douglas |author1-link=Douglas Lenat |chapter=On the thresholds of knowledge |title=Foundations Of Artificial Intelligence |editor1-last=Kirsh |editor1-first=David |publisher=MIT Press |pages=185–250 |year=1992}}</ref> An inference engine would have to be able to process huge numbers of rules to reach a decision.
-How to verify that decision rules are consistent with each other is also a challenge when there are too many rules. Usually such problem leads to a [[satisfiability]] (SAT) formulation.<ref>{{cite conference |vauthors=Bezem M |work=9th International Conference on Automated Deduction |title=Consistency of rule-based expert systems |volume=310 |date=1988 |pages=151–161 |doi=10.1007/BFb0012830 |isbn=3-540-19343-X |series=Lecture Notes in Computer Science |url=https://ir.cwi.nl/pub/6113|url-access=subscription }}</ref> This is a well-known NP-complete problem [[Boolean satisfiability problem]]. If we assume only [[binary data|binary variables]], say n of them, and then the corresponding search space is of size 2<math>^{n}</math>. Thus, the search space can grow exponentially.
+How to verify that decision rules are consistent with each other is also a challenge when there are too many rules. Usually such problem leads to a [[satisfiability]] (SAT) formulation.<ref>{{cite conference |vauthors=Bezem M |work=9th International Conference on Automated Deduction |chapter=Consistency of rule-based expert systems |volume=310 |date=1988 |pages=151–161 |doi=10.1007/BFb0012830 |isbn=3-540-19343-X |series=Lecture Notes in Computer Science |chapter-url=https://ir.cwi.nl/pub/6113 |chapter-url-access=subscription |title=Archived copy |access-date=2020-08-30 |archive-date=2020-10-23 |archive-url=https://web.archive.org/web/20201023223526/https://ir.cwi.nl/pub/6113 |url-status=live }}</ref> This is a well-known NP-complete problem [[Boolean satisfiability problem]]. If we assume only [[binary data|binary variables]], say n of them, and then the corresponding search space is of size 2<math>^{n}</math>. Thus, the search space can grow exponentially.
 There are also questions on how to prioritize the use of the rules to operate more efficiently, or how to resolve ambiguities (for instance, if there are too many else-if sub-structures within one rule) and so on.<ref>{{cite journal |vauthors=Mak B, Schmitt BH, and Lyytinen K |title=User participation in knowledge update of expert systems |journal=Information & Management |date=1997 |volume=32 |issue=2 |pages=55–63 |doi=10.1016/S0378-7206(96)00010-9 |doi-access=free}}</ref>
-Other problems are related to the [[overfitting]] and [[overgeneralization]] effects when using known facts and trying to generalize to other cases not described explicitly in the knowledge base. Such problems exist with methods that employ machine learning approaches too.<ref>{{cite book |vauthors=Pham HN, Triantaphyllou E |title=Soft Computing for Knowledge Discovery and Data Mining |chapter=The Impact of Overfitting and Overgeneralization on the Classification Accuracy in Data Mining |date=2008 |pages=391–431 |doi=10.1007/978-0-387-69935-6_16 |isbn=978-0-387-69934-9 |s2cid=12628921 |chapter-url=https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=4334&context=gradschool_dissertations}}</ref><ref>{{cite journal |vauthors=Pham HN, Triantaphyllou E |title=Prediction of diabetes by employing a new data mining approach which balances fitting and generalization |journal=Computer and Inf. Science G |date=2008 |pages=11–26}}</ref>
+Other problems are related to the [[overfitting]] and [[overgeneralization]] effects when using known facts and trying to generalize to other cases not described explicitly in the knowledge base. Such problems exist with methods that employ machine learning approaches too.<ref>{{cite book |vauthors=Pham HN, Triantaphyllou E |title=Soft Computing for Knowledge Discovery and Data Mining |chapter=The Impact of Overfitting and Overgeneralization on the Classification Accuracy in Data Mining |date=2008 |pages=391–431 |doi=10.1007/978-0-387-69935-6_16 |isbn=978-0-387-69934-9 |s2cid=12628921 |chapter-url=https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=4334&context=gradschool_dissertations |archive-date=2023-07-05 |access-date=2023-07-23 |archive-url=https://web.archive.org/web/20230705040238/https://digitalcommons.lsu.edu/cgi/viewcontent.cgi?article=4334&context=gradschool_dissertations |url-status=live }}</ref><ref>{{cite journal |vauthors=Pham HN, Triantaphyllou E |title=Prediction of diabetes by employing a new data mining approach which balances fitting and generalization |journal=Computer and Inf. Science G |date=2008 |pages=11–26}}</ref>
-Another problem related to the knowledge base is how to make updates of its knowledge quickly and effectively.<ref>{{cite journal |vauthors=Shan N, and Ziarko W |title=Data-based acquisition and incremental modification of classification rules |journal=Computational Intelligence |date=1995 |volume=11 |issue=2 |pages=357–370 |doi=10.1111/j.1467-8640.1995.tb00038.x |s2cid=38974914}}</ref><ref>{{cite journal |vauthors=Coats PK |title=Why expert systems fail |journal=Financial Management |volume=17 |issue=3 |date=1988 |pages=77–86 |jstor=3666074}}</ref><ref>{{cite journal |vauthors=Hendriks PH, and Vriens DJ |title=Knowledge-based systems and knowledge management: friends or foes? |journal=Information & Management |date=1999 |volume=35 |issue=2 |pages=113–125 |doi=10.1016/S0378-7206(98)00080-9 |url=https://repository.ubn.ru.nl/handle/2066/240554|hdl=2066/240554 |hdl-access=free }}</ref> Also how to add a new piece of knowledge (i.e., where to add it among many rules) is challenging. Modern approaches that rely on machine learning methods are easier in this regard.{{Citation needed|date=October 2019}}
+Another problem related to the knowledge base is how to make updates of its knowledge quickly and effectively.<ref>{{cite journal |vauthors=Shan N, and Ziarko W |title=Data-based acquisition and incremental modification of classification rules |journal=Computational Intelligence |date=1995 |volume=11 |issue=2 |pages=357–370 |doi=10.1111/j.1467-8640.1995.tb00038.x |s2cid=38974914}}</ref><ref>{{cite journal |vauthors=Coats PK |title=Why expert systems fail |journal=Financial Management |volume=17 |issue=3 |date=1988 |pages=77–86 |jstor=3666074}}</ref><ref>{{cite journal |vauthors=Hendriks PH, and Vriens DJ |title=Knowledge-based systems and knowledge management: friends or foes? |journal=Information & Management |date=1999 |volume=35 |issue=2 |pages=113–125 |doi=10.1016/S0378-7206(98)00080-9 |url=https://repository.ubn.ru.nl/handle/2066/240554 |hdl=2066/240554 |hdl-access=free |archive-date=2024-04-19 |access-date=2024-02-20 |archive-url=https://web.archive.org/web/20240419185339/https://repository.ubn.ru.nl/handle/2066/240554 |url-status=live }}</ref> Also how to add a new piece of knowledge (i.e., where to add it among many rules) is challenging. Modern approaches that rely on machine learning methods are easier in this regard.{{Citation needed|date=October 2019}}
 Because of the above challenges, it became clear that new approaches to AI were required instead of rule-based technologies. These new approaches are based on the use of machine learning techniques, along with the use of feedback mechanisms.<ref name="CADsurvey"/>
-The key challenges that expert systems in medicine (if one considers computer-aided diagnostic systems as modern expert systems), and perhaps in other application domains, include issues related to aspects such as: big data, existing regulations, healthcare practice, various algorithmic issues, and system assessment.<ref>{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=The Seven Key Challenges for the Future of Computer-Aided Diagnosis in Medicine |journal=International Journal of Medical Informatics |volume=129 |pages=413–422 |date=2019 |doi=10.1016/j.ijmedinf.2019.06.017 |pmid=31445285 |s2cid=198287435}}</ref>
+The key challenges that expert systems in medicine (if one considers computer-aided diagnostic systems as modern expert systems), and perhaps in other application domains, include issues related to aspects such as: big data, existing regulations, healthcare practice, various algorithmic issues, and system assessment.<ref>{{cite journal |vauthors=Yanase J, Triantaphyllou E |title=The Seven Key Challenges for the Future of Computer-Aided Diagnosis in Medicine |journal=International Journal of Medical Informatics |volume=129 |pages=413–422 |date=2019 |doi=10.1016/j.ijmedinf.2019.06.017 |pmid=31445285 |s2cid=198287435 |url=https://repository.lsu.edu/eecs_pubs/1347 |archive-date=2025-08-13 |access-date=2026-06-01 |archive-url=https://web.archive.org/web/20250813104551/https://repository.lsu.edu/eecs_pubs/1347/ |url-status=live }}</ref>
 Finally, the following disadvantages of using expert systems can be summarized:<ref name=":0"/>
@@ Line 120: / Line 122: @@
 | Prediction || Inferring likely consequences of given situations || Preterm Birth Risk Assessment<ref>{{cite journal |last1=Woolery |first1=L.K. |last2=Grzymala-Busse |first2=J. |year=1994 |title=Machine learning for an expert system to predict preterm birth risk |journal=Journal of the American Medical Informatics Association |volume=1 |issue=6 |pages=439–446 |pmc=116227 |pmid=7850569 |doi=10.1136/jamia.1994.95153433}}</ref>
 |-
-| Diagnosis || Inferring system malfunctions from observables || [[CADUCEUS (expert system)|CADUCEUS]], [[MYCIN]], PUFF, Mistral,<ref name="Mistral">{{cite journal |last1=Salvaneschi |first1=Paolo |first2=Mauro |last2=Cadei |first3=Marco |last3=Lazzari |year=1996 |title=Applying AI to structural safety monitoring and evaluation |journal=IEEE Expert |volume=11 |issue=4 |pages=24–34 |url=http://www.computer.org/csdl/mags/ex/1996/04/x4024-abs.html|access-date=5 March 2014 |doi=10.1109/64.511774 |url-access=subscription}}</ref> Eydenet,<ref name="Eydenet"/> Kaleidos,<ref name="Kaleidos"/> GARVAN-ES1<ref name="GARVAN">{{cite journal |author1=K. Horn |author2=L. Lazarus |author3=P. Compton |author4=J.R. Quinlan |title=An expert system for the interpretation of thyroid assays in a clinical laboratory |journal=Aust Comput J |year=1985 |volume=17 |pages=7–11}}</ref><ref name="Buchanan">{{cite journal |author=Buchanan, B. |title=Expert systems: working systems and the research literature |journal=Expert Systems |year=1986 |volume=3 |issue=1 |pages=32–51 |doi= 10.1111/j.1468-0394.1986.tb00192.x}}</ref><ref name="Maintenance">{{cite journal |author1=P. Compton |author2=K. Horn |author3=R. Quinlan |author4=L. Lazarus |author5=K. Ho |title=Maintaining an Expert System |journal=Proceedings of the Fourth Australian Conference on the Applications of Expert Systems |year=1988}}</ref>
+| Diagnosis || Inferring system malfunctions from observables || [[CADUCEUS (expert system)|CADUCEUS]], [[MYCIN]], PUFF, Mistral,<ref name="Mistral">{{cite journal |last1=Salvaneschi |first1=Paolo |first2=Mauro |last2=Cadei |first3=Marco |last3=Lazzari |year=1996 |title=Applying AI to structural safety monitoring and evaluation |journal=IEEE Expert |volume=11 |issue=4 |pages=24–34 |url=http://www.computer.org/csdl/mags/ex/1996/04/x4024-abs.html |access-date=5 March 2014 |doi=10.1109/64.511774 |bibcode=1996IExp...11...24S |url-access=subscription |archive-date=5 March 2014 |archive-url=https://web.archive.org/web/20140305053153/http://www.computer.org/csdl/mags/ex/1996/04/x4024-abs.html |url-status=live }}</ref> Eydenet,<ref name="Eydenet"/> Kaleidos,<ref name="Kaleidos"/> GARVAN-ES1<ref name="GARVAN">{{cite journal |author1=K. Horn |author2=L. Lazarus |author3=P. Compton |author4=J.R. Quinlan |title=An expert system for the interpretation of thyroid assays in a clinical laboratory |journal=Aust Comput J |year=1985 |volume=17 |pages=7–11}}</ref><ref name="Buchanan">{{cite journal |author=Buchanan, B. |title=Expert systems: working systems and the research literature |journal=Expert Systems |year=1986 |volume=3 |issue=1 |pages=32–51 |doi= 10.1111/j.1468-0394.1986.tb00192.x}}</ref><ref name="Maintenance">{{cite journal |author1=P. Compton |author2=K. Horn |author3=R. Quinlan |author4=L. Lazarus |author5=K. Ho |title=Maintaining an Expert System |journal=Proceedings of the Fourth Australian Conference on the Applications of Expert Systems |year=1988}}</ref>
 |-
-| Design || Configuring objects under constraints || [[Dendral]], [[Expert systems for mortgages|Mortgage Loan Advisor]], [[Xcon|R1]] (DEC VAX Configuration), SID (DEC [[VAX 9000]] [[central processing unit|CPU]])
+| Design || Configuring objects under constraints || [[Dendral]], [[Expert systems for mortgages|Mortgage Loan Advisor]], [[Xcon|R1]] (DEC VAX Configuration), SID (DEC [[VAX 9000]] [[central processing unit|CPU]]), Database Design Advisor<ref>{{cite conference
+ |chapter=Design and implementation of a database design aid using VP-Expert
+ |last1=Song
+ |first1=Il-Yeol
+ |last2=Strum
+ |first2=S. D.
+ |last3=Medsker
+ |first3=Carl J.
+ |title=&#91;1991&#93; Proceedings of the IEEE/ACM International Conference on Developing and Managing Expert System Programs
+ |year=1991
+ |pages=15–23
+ |publisher=IEEE
+ |doi=10.1109/DMESP.1991.171764
+ |isbn=0-8186-2250-4
+ |chapter-url=https://www.computer.org/csdl/proceedings-article/dmesp/1991/00171764/12OmNrkT7EE
+ |access-date=2025-09-28
+ |archive-date=2024-12-15
+ |archive-url=https://web.archive.org/web/20241215023737/https://www.computer.org/csdl/proceedings-article/dmesp/1991/00171764/12OmNrkT7EE
+ |url-status=live
+ }}</ref>
 |-
 | [[Automated planning and scheduling|Planning]] || Designing actions || Mission Planning for Autonomous Underwater Vehicle<ref>{{cite book |last=Kwak |first=S. H. |title=Symposium on Autonomous Underwater Vehicle Technology |chapter=A mission planning expert system for an autonomous underwater vehicle |year=1990 |pages=123–128|doi=10.1109/AUV.1990.110446 |s2cid=60476847}}</ref>
 |-
-| Monitoring || Comparing observations to plan vulnerabilities || REACTOR<ref>{{cite journal |last=Nelson |first=W. R. |title=REACTOR: An Expert System for Diagnosis and Treatment of Nuclear Reactors |url=https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4daf79217bb76482943f07b5138104a45f85c076 |year=1982 |journal=AAAI |volume=82: Proceedings of the Second AAAI Conference on Artificial Intelligence |pages=296–301}}</ref>
+| Monitoring || Comparing observations to plan vulnerabilities || REACTOR<ref>{{cite journal |last=Nelson |first=W. R. |title=REACTOR: An Expert System for Diagnosis and Treatment of Nuclear Reactors |url=https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4daf79217bb76482943f07b5138104a45f85c076 |year=1982 |journal=AAAI |volume=82: Proceedings of the Second AAAI Conference on Artificial Intelligence |pages=296–301 |archive-date=2023-03-11 |access-date=2023-03-11 |archive-url=https://web.archive.org/web/20230311174105/https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4daf79217bb76482943f07b5138104a45f85c076 |url-status=live }}</ref>
 |-
 | Debugging || Providing incremental solutions for complex problems || SAINT, MATHLAB, MACSYMA
@@ Line 134: / Line 156: @@
 | Instruction || Diagnosing, assessing, and correcting student behaviour || SMH.PAL,<ref name="smhpal"/> Intelligent Clinical Training,<ref>{{cite journal |last1=Haddawy |first1=P |first2=S. |last2=Suebnukarn |year=2010 |title=Intelligent Clinical Training Systems |journal=Methods Inf Med |volume=49 |issue=4 |pages=388–9 |citeseerx=10.1.1.172.60 |pmid=20686730|doi=10.1055/s-0038-1625342 |s2cid=11903941}}</ref> STEAMER<ref>{{cite journal |last1=Hollan |first1=J. |first2=E. |last2=Hutchins |first3=L. |last3=Weitzman |year=1984 |title=STEAMER: An interactive inspectable simulation-based training system |journal=AI Magazine}}</ref>
 |-
 | Control || Interpreting, predicting, repairing, and monitoring system behaviors || Real Time Process Control,<ref>{{cite journal |last=Stanley |first=G.M. |title=Experience Using Knowledge-Based Reasoning in Real Time Process Control |journal=Plenary Paper Presented at: International Federation of Automatic Control (IFAC) Symposium on Compute R Aided Design in Control Systems|date=July 15–17, 1991|url=http://www.gregstanleyandassociates.com/whitepapers/IFAC91objectPaper.pdf|access-date=3 December 2013}}</ref> Space Shuttle Mission Control,<ref>{{cite journal |last1=Rasmussen |first1=Arthur |first2=John F. |last2=Muratore |first3=Troy A. |last3=Heindel |title=The INCO Expert System Project: CLIPS in Shuttle mission control |journal=NTRS|date=February 1990|url=https://www.researchgate.net/publication/4702412|access-date=30 November 2013}}</ref> Smart Autoclave Cure of Composites<ref>{{Cite book |author1=Ciriscioli, P. R. |author2=G. S. Springer |title="Smart Autoclave Cure of Composites" |year=1990 |isbn=9781003209010}}</ref>
 |}
@@ Line 147: / Line 170: @@
 GARVAN-ES1 was a medical expert system, developed at the [[Garvan Institute of Medical Research]], that provided automated clinical diagnostic comments on endocrine reports from a pathology laboratory. It was one of the first medical expert systems to go into routine clinical use internationally<ref name="Buchanan"/> and the first expert system to be used for diagnosis daily in Australia.<ref name="Catlett">{{cite journal |author=Catlett, J. |title=Expert systems, the risks and rewards |journal=Hub Information Technology |year=1990 |volume=2 |issue=7 |pages=20–26}}</ref> The system was written in "C" and ran on a PDP-11 in 64K of memory. It had 661 rules that were compiled; not interpreted.
-Mistral<ref name="Mistral"/> is an expert system to monitor dam safety, developed in the 1990s by Ismes (Italy). It gets data from an automatic monitoring system and performs a diagnosis of the state of the dam. Its first copy, installed in 1992 on the [[Ridracoli]] Dam (Italy), is still operational 24/7/365. It has been installed on several dams in Italy and abroad (e.g., [[Itaipu Dam]] in Brazil), and on landslide sites under the name of Eydenet,<ref name="Eydenet">{{cite journal |last1=Lazzari |first1=Marco |first2=Paolo |last2=Salvaneschi |title=Embedding a geographic information system in a decision support system for landslide hazard monitoring |journal=International Journal of Natural Hazards |year=1999 |volume=20 |issue=2–3 |pages=185–195|url=http://dinamico2.unibg.it/lazzari/doc/embedding-authors-copy.pdf|doi=10.1023/A:1008187024768 |bibcode=1999NatHa..20..185L |s2cid=1746570}}</ref> and on monuments under the name of Kaleidos.<ref name="Kaleidos">{{cite journal |last1=Lancini |first1=Stefano |first2=Marco |last2=Lazzari |first3=Alberto |last3=Masera |first4=Paolo |last4=Salvaneschi |title=Diagnosing Ancient Monuments with Expert Software |journal=Structural Engineering International |year=1997 |volume=7 |issue=4 |pages=288–291|url=http://dinamico2.unibg.it/lazzari/doc/structural-engineering-authors-copy.pdf|doi=10.2749/101686697780494392 |s2cid=8113173}}</ref> Mistral is a registered trade mark of [[Centro Elettrotecnico Sperimentale Italiano|CESI]].{{Missing information|section|which date ''"is still"'' in ''"Its first copy, installed in 1992 on the Dam (Italy), is still operational 24/7/365."'' refers to, all the sources cited seem to be from before 2000|date=January 2022}}
+Mistral<ref name="Mistral"/> is an expert system to monitor dam safety, developed in the 1990s by Ismes in Italy. It gets data from an automatic monitoring system and performs a diagnosis of the state of the dam. Its first copy was installed in 1992 on the [[Ridracoli]] Dam in Italy. It has been installed on several other dams in Italy and abroad (e.g., [[Itaipu Dam]] in Brazil), and on landslide sites under the name of Eydenet,<ref name="Eydenet">{{cite journal |last1=Lazzari |first1=Marco |first2=Paolo |last2=Salvaneschi |title=Embedding a geographic information system in a decision support system for landslide hazard monitoring |journal=International Journal of Natural Hazards |year=1999 |volume=20 |issue=2–3 |pages=185–195 |url=http://dinamico2.unibg.it/lazzari/doc/embedding-authors-copy.pdf |doi=10.1023/A:1008187024768 |bibcode=1999NatHa..20..185L |s2cid=1746570 |archive-date=2014-10-17 |access-date=2014-10-12 |archive-url=https://web.archive.org/web/20141017145920/http://dinamico2.unibg.it/lazzari/doc/embedding-authors-copy.pdf |url-status=live }}</ref> and on monuments under the name of Kaleidos.<ref name="Kaleidos">{{cite journal |last1=Lancini |first1=Stefano |first2=Marco |last2=Lazzari |first3=Alberto |last3=Masera |first4=Paolo |last4=Salvaneschi |title=Diagnosing Ancient Monuments with Expert Software |journal=Structural Engineering International |year=1997 |volume=7 |issue=4 |pages=288–291 |url=http://dinamico2.unibg.it/lazzari/doc/structural-engineering-authors-copy.pdf |doi=10.2749/101686697780494392 |s2cid=8113173 |archive-date=2015-04-13 |access-date=2014-10-12 |archive-url=https://web.archive.org/web/20150413032053/http://dinamico2.unibg.it/lazzari/doc/structural-engineering-authors-copy.pdf |url-status=live }}</ref> Mistral is a registered trade mark of [[Centro Elettrotecnico Sperimentale Italiano|CESI]].
 == See also ==
 * [[AI winter]]

Expert system: Difference between revisions

Navigation menu

Search