Expert System Participants Domain expert Knowledge engineer Knowledge user
The domain expert The individual or group that has the expertise or knowledge one is trying to capture in the expert system. Recognizes the real problem develops a general framework for problem solving formulates theories about the situation develops and uses general rules to solve a problem knows when to break the rules or general principles solves problems quickly and efficiently learns from past experiences knows what is important in solving a problem and what is not explains the situation and solutions of problems to others
The knowledge engineer The individual who has training and/or experience in the design, development, implementation, and maintenance of an expert system, including training and/or experience with expert system shells. As expert systems shells and development tools become easier to use, the role of the knowledge engineer shifts from system developer to consultant
The Knowledge User The individual or group who uses and benefits from the expert system. Knowledge users do not need any previous training in computers or expert systems.
Expert System Development Theoretically, expert systems can be developed from any programming language. Early days: Pascal, FORTRAN, COBOL NEXT: LISP for AI, then Prolog for AI Now: Financial Advisor, 1st Class Fusion, Knowledepro, Leonardo, Personal Consultant (PC) Easy are a few used today.
Other Business Uses of Expert Systems Kaiser Permanente- uses and expert system and voice response to automate its help desk function Nynex(newyork/new england tel exch)--uses expert systems to help its workers locate and solve customer-related phone problems MYCIN(Stanford) uses system to analyze blood infections CARGEX-uses system to help determine the best shipping routes CoverStory-uses system to extract marketing info from database and automatically write marketing reports.
Integrating Expert Systems An expert system can be integrated with other information systems in an organization through a common database. Expert systems support managerial decision making and organizational goals.
Think About it Think of five ideas for expert systems that would be useful on campus Try to develop rules you might use to make decisions (ie: to choose classes for a quarter) this exercise will show the difficulty of using multiple experts (other students in your group) to build a knowledge base and the difficulty of defining rules for a knowledge base.