Expert Systems Apps

By pjain      Published Aug. 28, 2020, 4:55 a.m. in blog AI-Analytics-Data   

ES Algorithms -- xfr --

Custom Inference Systems, Rules Processing

Rules Engine Algorithms Processing

Expert Systems Core Algorithms

General Purpose Shells

Knowledge Engineering Based Approaches

Expert Systems Apps


  • Expert system - Wikipedia
  • "Expert systems: perils and promise", D. G. Bobrow, S. Mittal, M. J. Stefik. Communications of the ACM, pp 880 - 894, issue 9, volume 29, (September 1986)
  • The AI Business: The commercial uses of artificial intelligence, ed. Patrick Winston and Karen A. Prendergast. 1984. ISBN 0-262-23117-4

Finance and Fraud Detection

Why AI in Finance Apps

  • Time to Approval - between the first contact with the customer and the bank's offering of a loan - the best high quality loan customers want fast closes, and are not willing to wait 20 or more days. So if a bank's processing is slower than competition, it will lose out.

  • Quality of a mortgage loans portfolio to the bank

  • AI/expert systems can capitalize on regulatory possibilities and necessities.

Mortgage Underwriting

After 2008 GFC, many of global mortgage markets are subject to government regulation as well as future underwriting and guarantees or buying as part of bundles and securitization (eg FHA/VA, Ginnie/Freddi-mac, etc).

Users are shopping for best rates AND Closing Fees. In a competitive environment, that leads to process operational efficiency being required of a mortgage process. While jumbo mortgages eg $750,000 and up in the coasts may support full process, yet often the average house price in USA is $250,000.

Loan departments are interested in expert systems for mortgages because of the growing cost of labor which makes the handling and acceptance of relatively small loans less profitable. They also see in the application of expert systems a possibility for standardized, efficient handling of mortgage loans, and appreciate that for the acceptance of mortgages there are hard and fast rules which do not always exist with other types of loans.

Special Cases, programs for Affordability need LOTS of work - AI can help

In France, the government subsidizes one type of loan which is available only on low-cost properties (the HLM) and to lower income families. Known as "frets Conventionnes", these carry a rate of interest lower than the rate on the ordinary property loan from a bank. The difficulty is that granting them is subject to numerous regulations, concerning both:

the home which is to be purchased, and
the financial circumstances of the borrower.

To assure that all conditions have been met, every application has to be first processed at branch level and then sent to a central office for checking, before going back to the branch, often with requests for more information from the applicant. This leads to frustrating delays. Expert system for mortgages takes care of these by providing branch employees with tools permitting them to process an application correctly, even if a bank employee does not have an exact knowledge of the screening procedure.

Insurance Industry Applications

Car Body Damage Claims

Car body damage which focus on panels and Bill-of-Materials, so adjustment can be handled fairly routinely.

Adjusters in the evaluation of bodily injury claims

Colossus, a computer program, developed by Computer Sciences Corporation is the insurance industry’s leading expert system for assisting adjusters in the evaluation of bodily injury claims (aka "pain and suffering"). Colossus helps adjusters reduce variance in payouts on similar bodily injury claims through objective use of industry standard rules.


  • Expert systems for mortgages - Wikipedia
  • Steinmann, Heinrich; Chorafas, Dimitris N. (1990). Expert systems in banking: a guide for senior managers. New York: New York University Press. pp. 222–225. ISBN 0-8147-1449-8.
  • Mishler, Lon; Cole, Robert E. (1995). Consumer and business credit management. Homewood, Ill: Irwin. p. 115. ISBN 0-256-13948-2.
  • Clancy, Paul, Gerald Hoenig, and Arnold Schmitt. 1989. An Expert System for Legal Consultation. In Proceedings of the Second Annual Conference on Innovative Applications of Artificial Intelligence, 125 - 135. Menlo Park, Calif.: AAAI Press.

Investing and Algorithmic Trading Apps

HR Apps

Medical Apps

Diagnostic Apps

From the earliest days of computers biomedical apps have been a priority for computer applications

Early diagnostic systems used patients’ symptoms and laboratory test results as inputs to generate a diagnostic outcome. However, researchers had realized that there were significant limitations when using traditional methods like flow-charts, statistical pattern-matching, or probability theory.

  • Ledley RS, and Lusted LB (1959). "Reasoning foundations of medical diagnosis". Science. 130 (3366): 9–21. Bibcode:1959Sci...130....9L. doi:10.1126/science.130.3366.9
  • Weiss SM, Kulikowski CA, Amarel S, Safir A (1978). "A model-based method for computer-aided medical decision-making". Artificial Intelligence. 11 (1–2): 145–172. doi:10.1016/0004-3702(78)90015-2


  • Schwartz WB (1970). "Medicine and the computer: the promise and problems of change". New England Journal of Medicine. 283 (23): 1257–1264. doi:10.1056/NEJM197012032832305
  • Bleich HL (1972). "Computer-based consultation: Electrolyte and acid-base disorders". The American Journal of Medicine. 53 (3): 285–291. doi:10.1016/0002-9343(72)90170-2

Statistical pattern-matching

  • Rosati RA, McNeer JF, Starmer CF, Mittler BS, Morris JJ, and Wallace AG (1975). "A new information system for medical practice". Archives of Internal Medicine. 135 (8): 1017–1024. doi:10.1001/archinte.1975.00330080019003

Probability theory

  • Gorry GA, Kassirer JP, Essig A, and Schwartz WB (1973). "Decision analysis as the basis for computer-aided management of acute renal failure". The American Journal of Medicine. 55 (4): 473–484. doi:10.1016/0002-9343(73)90204-0
  • Szolovits P, Patil RS, and Schwartz WB (1988). "Artificial intelligence in medical diagnosis". Annals of Internal Medicine. 108 (1): 80–87. doi:10.7326/0003-4819-108-1-80

DENDRAL project at Stanford, 1973

  • 15 years ago, with its objective of problem-solving via the automation of actual human expert knowledge

MYCIN, Stanford

MYCIN was an early backward chaining expert system that used artificial intelligence to identify bacteria causing severe infections, such as bacteremia and meningitis, and to recommend antibiotics, with the dosage adjusted for patient's body weight — the name derived from the antibiotics themselves, as many antibiotics have the suffix "-mycin". The Mycin system was also used for the diagnosis of blood clotting diseases. MYCIN was developed over five or six years in the early 1970s at Stanford University. It was written in Lisp as the doctoral dissertation of Edward Shortliffe under the direction of Bruce G. Buchanan, Stanley N. Cohen and others.

  • The original MYCIN domain rules were inaccessible control knowledge implicit in the system
  • NEOMYCIN program
  • Metarules in TEIRESIAS (a knowledge acquisition and explanation facility for MYCIN)

  • Shortliffe EH, and Buchanan BG (1975). "A model of inexact reasoning in medicine". Mathematical Biosciences. 23 (3–4): 351–379. doi:10.1016/0025-5564(75)90047-4

INTERNIST-I expert system

  • Miller RA, Pople Jr HE, and Myers JD (1982). "Internist-I, an experimental computer-based diagnostic consultant for general internal medicine". New England Journal of Medicine. 307 (8): 468–476. doi:10.1056/NEJM198208193070803
  • First, MB; Soffer, LJ; Miller, RA (1985). "QUICK (QUick Index to Caduceus Knowledge): using the INTERNIST-1/CADUCEUS knowledge base as an electronic textbook of medicine". Computers and Biomedical Research. 18 (2): 137–65. doi:10.1016/0010-4809(85)90041-2. PMID 3886276.


In the mid 1980s, CADUCEUS was a medical expert system finished in the mid-1980s (first begun in the 1970s- it took a long time to build the knowledge base) by Harry Pople (of the University of Pittsburgh), building on Pople's years of interviews with Dr. Jack Meyers, one of the top internal medicine diagnosticians and a professor at the University of Pittsburgh. Their motivation was an intent to improve on MYCIN (which focused on blood-borne infectious bacteria) to focus on more comprehensive issues than a narrow field like blood poisoning (though it would do it in a similar manner); instead embracing all internal medicine. CADUCEUS eventually could diagnose up to 1000 different diseases.

While CADUCEUS worked using an inference engine similar to MYCIN's, it made a number of changes (like incorporating abductive reasoning) to deal with the additional complexity of internal disease- there can be a number of simultaneous diseases, and data is generally flawed and scarce.

CADUCEUS has been described as the "most knowledge-intensive expert system in existence".

  • Banks, G (1986). "Artificial intelligence in medical diagnosis: the INTERNIST/CADUCEUS approach". Critical Reviews in Medical Informatics. 1 (1): 23–54. PMID 3331578.
  • Wolfram, D (1995). "An appraisal of INTERNIST-I". Artificial Intelligence in Medicine. 7 (2): 93–116. doi:10.1016/0933-3657(94)00028-Q. PMID 7647840.
  • First, MB; Soffer, LJ; Miller, RA (1985). "QUICK (QUick Index to Caduceus Knowledge): using the INTERNIST-1/CADUCEUS knowledge base as an electronic textbook of medicine". Computers and Biomedical Research. 18 (2): 137–65. doi:10.1016/0010-4809(85)90041-2. PMID 3886276.


There are no comments yet

Add new comment

Similar posts

App Store Marketing

tr Expert Systems Papers

mr ES Frameworks Review

Expert Systems Architecture