AI Knowledge-Based Decision Support Expert Systems

2 Uses of Knowledge Knowledge consists of facts, concepts, theories, heuristic methods, procedures, and relationships Knowledge is also information organized and analyzed for understanding and applicable to problem solving or decision making Knowledge base - the collection of knowledge related to a problem (or opportunity) used in an AI system Typically limited in some specific, usually narrow, subject area or domain The narrow domain of knowledge, and that an AI system must involve some qualitative aspects of decision making (critical for AI application success) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3 Knowledge Bases Search the Knowledge Base for Relevant Facts and Relationships Reach One or More Alternative Solutions to a Problem Augments the User (Typically a Novice) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4 4 (ES) Introduction n Expert System vs. knowledge-based system n An Expert System is a system that employs human knowledge captured in a computer to solve problems that ordinarily require human expertise n ES imitate the expert’s reasoning processes to solve specific problems

5 5 History of Expert Systems 1. Early to Mid-1960s – One attempt: the General-purpose Problem Solver (GPS) n General-purpose Problem Solver (GPS) n A procedure developed by Newell and Simon [1973] from their Logic Theory Machine - – Attempted to create an "intelligent" computer general problem-solving methods applicable across domains – Predecessor to ES – Not successful, but a good start

Mid-1960s: Special-purpose ES programs – DENDRAL – MYCIN n Researchers recognized that the problem- solving mechanism is only a small part of a complete, intelligent computer system – General problem solvers cannot be used to build high performance ES – Human problem solvers are good only if they operate in a very narrow domain – Expert systems must be constantly updated with new information – The complexity of problems requires a considerable amount of knowledge about the problem area

Mid 1970s – Several Real Expert Systems Emerge – Recognition of the Central Role of Knowledge – AI Scientists Develop Comprehensive knowledge representation theories General-purpose, decision-making procedures and inferences n Limited Success Because – Knowledge is Too Broad and Diverse – Efforts to Solve Fairly General Knowledge- Based Problems were Premature

8 8 BUT n Several knowledge representations worked Key Insight n The power of an ES is derived from the specific knowledge it possesses, not from the particular formalisms and inference schemes it employs

Early 1980s n ES Technology Starts to go Commercial – XCON – XSEL – CATS-1 n Programming Tools and Shells Appear – EMYCIN – EXPERT – META-DENDRAL – EURISKO n About 1/3 of These Systems Are Very Successful and Are Still in Use

10 Latest ES Developments n Many tools to expedite the construction of ES at a reduced cost n Dissemination of ES in thousands of organizations n Extensive integration of ES with other CBIS n Increased use of expert systems in many tasks n Use of ES technology to expedite IS construction ( ES Shell )

11 n The object-oriented programming approach in knowledge representation n Complex systems with multiple knowledge sources, multiple lines of reasoning, and fuzzy information n Use of multiple knowledge bases n Improvements in knowledge acquisition n Larger storage and faster processing computers n The Internet to disseminate software and expertise.

12 Expert Systems Attempt to Imitate Expert Reasoning Processes and Knowledge in Solving Specific Problems Most Popular Applied AI Technology –Enhance Productivity –Augment Work Forces Narrow Problem-Solving Areas or Tasks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

13 Expert Systems Provide Direct Application of Expertise Expert Systems Do Not Replace Experts, But They –Make their Knowledge and Experience More Widely Available –Permit Nonexperts to Work Better Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

14 Expertise The extensive, task-specific knowledge acquired from training, reading and experience –Theories about the problem area –Hard-and-fast rules and procedures –Rules (heuristics) –Global strategies –Meta-knowledge (knowledge about knowledge) –Facts Enables experts to be better and faster than nonexperts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

15 Human Expert Behaviors Recognize and formulate the problem Solve problems quickly and properly Explain the solution Learn from experience Restructure knowledge Break rules Determine relevance Degrade gracefully Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

16 Transferring Expertise Objective of an expert system –To transfer expertise from an expert to a computer system and –Then on to other humans (nonexperts) Activities –Knowledge acquisition –Knowledge representation –Knowledge inferencing –Knowledge transfer to the user Knowledge is stored in a knowledge base Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

17 Inferencing Reasoning (Thinking) The computer is programmed so that it can make inferences Performed by the Inference Engine Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

18 Rules IF-THEN-ELSE Explanation Capability –By the justifier, or explanation subsystem ES versus Conventional Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

19 22 Knowledge as Rules n MYCIN rule example: IF the infection is meningitis AND patient has evidence of serious skin or soft tissue infection AND organisms were not seen on the stain of the culture AND type of infection is bacterial THEN There is evidence that the organism (other than those seen on cultures or smears) causin the infection is Staphylococus coagpus.

20 Three Major ES Components User Interface Inference Engine Knowledge Base

21 26 Knowledge Base Inference Engine User Interface Explanation Facility Working Memory Knowledge Acquisition ES Shell Database, Spreadsheets, etc. Basic ES Structure

22 All ES Components Knowledge Acquisition Subsystem Knowledge Base Inference Engine User Interface Blackboard (Workplace) Explanation Subsystem (Justifier) Knowledge Refining System User Most ES do not have a Knowledge Refinement Component (See Figure 10.3)

And now with confidences n Facts: –F1: Ungee gives milk:.9 –F2: Ungee eats meat:.8 – F3: Ungee has hoofs:.7 n Rules: –R1: If X gives milk, then it is a mammal:.6 –R2: If X is a mammal and eats meat, then carnivore:.5 –R3: If X has hoofs, then X is carnivore:.4

n R1 with F1: Ungee is mammal. (F4) n Confidence F4: C(F4) =.9*.6 =.54 n R2 using F2 and F4 yields: Ungee is carnivore (F5). n C(F5) from R2 = min(.54,.8)*.5 =.27 n R3 using F3 conclude F5 from R3 n C(F5) from R3 =.7*.4 =.28 n C(F5) from R3 and R2 = 1 –(1-.28)*(1-.27) =.48

25 Knowledge Base The knowledge base contains the knowledge necessary for understanding, formulating, and solving problems Two Basic Knowledge Base Elements –Facts –Special heuristics, or rules that direct the use of knowledge –Knowledge is the primary raw material of ES –Incorporated knowledge representation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

26 Inference Engine The brain of the ES The control structure (rule interpreter) Provides methodology for reasoning

The production system after Rule 1 has fired.

The system after Rule 4 has fired. Note the stack- based approach to goal reduction.

The and/or graph searched in the car diagnosis example, with the conclusion of Rule 4 matching the first premise of Rule 1.

Explanation and Transparency in Goal-Driven Reasoning n The following dialogue begins with the computer asking the user about the goals present in the working memory: –Gas in fuel tank? YES –Gas in carburetor? YES –Engine will turn over? WHY

Model-Based Expert System Example n The expected output value are given in () and the actual outputs in [ ]

32 35 The Human Element in Expert Systems n Builder and User n Expert and Knowledge engineer. n The Expert – Has the special knowledge, judgment, experience and methods to give advice and solve problems – Provides knowledge about task performance

33 36 The Knowledge Engineer n Helps the expert(s) structure the problem area by interpreting and integrating human answers to questions, drawing analogies, posing counterexamples, and bringing to light conceptual difficulties n Usually also the System Builder

34 37 The User n Possible Classes of Users – A non-expert client seeking direct advice - the ES acts as a Consultant or Advisor – A student who wants to learn - an Instructor – An ES builder improving or increasing the knowledge base - a Partner – An expert - a Colleague or Assistant n The Expert and the Knowledge Engineer Should Anticipate Users' Needs and Limitations When Designing ES

35 Problem Areas Addressed by Expert Systems Interpretation systems Prediction systems Diagnostic systems Design systems Planning systems Monitoring systems Debugging systems Repair systems Instruction systems Control systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

36 Expert Systems Benefits Improved Decision Quality Increased Output and Productivity Decreased Decision Making Time Increased Process(es) and Product Quality Capture Scarce Expertise Can Work with Incomplete or Uncertain Information Enhancement of Problem Solving and Decision Making Improved Decision Making Processes Knowledge Transfer to Remote Locations Enhancement of Other MIS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

37 Problems and Limitations of Expert Systems Knowledge is not always readily available Expertise can be hard to extract from humans Expert system users have natural cognitive limits ES work well only in a narrow domain of knowledge Knowledge engineers are rare and expensive Lack of trust by end-users ES may not be able to arrive at valid conclusions ES sometimes produce incorrect recommendations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

38 Expert System Success Factors Most Critical Factors –Champion in Management –User Involvement and Training Plus –The level of knowledge must be sufficiently high –There must be (at least) one cooperative expert –The problem must be qualitative (fuzzy), not quantitative –The problem must be sufficiently narrow in scope –The ES shell must be high quality, and naturally store and manipulate the knowledge –A friendly user interface –Important and difficult enough problem