Week 6 Expert System Case Scenario During ABC

Week 6 Expert System

Case Scenario During ABC Enterprise management meeting to discuss whether the company should consider a merger with other business partners, one important key person of the company, Mr. Amir, who is in charge of the company’s investment portfolios was absent due to fever. The CEO of the company, Mr. Ridzuan wishes to make his expertise in investment portfolio to be made available whenever it is needed, especially during crucial time. Thus, Mr. Ridzuan was considering whether a system can be built to provide direct expertise in the same way that Mr. Amir provides investment portfolio consultation, such as type of investment, their rules, calculations, and so on.

Expert Systems n n Attempt to imitate expert reasoning processes and knowledge in solving specific problems. Most popular applied AI technology n n n Enhance Productivity Increase (value) Workforces Narrow problem-solving areas or tasks 3

Expert Systems n n Provide direct application of expertise Expert systems do not replace experts, but they n n Make their knowledge and experience more widely available Permit non-experts to work better 4

Expertise n n The extensive, task-specific knowledge acquired from training, reading and experience n Theories about the problem area n Hard-and-fast rules and procedures n Rules (heuristics) n Global strategies n Facts Enables experts to be better and faster than non-experts. 5

Human Expert Behaviors n n n n Recognize and formulate the problem Solve problems quickly and properly Explain the solution Learn from experience Restructure knowledge Break rules Determine relevance Degrade gracefully 6

Transferring Expertise n n n Objective of an expert system n To transfer expertise from an expert to a computer system and n Then on to other humans (non-experts) Activities n Knowledge acquisition n Knowledge representation n Knowledge inferencing n Knowledge transfer to the user Knowledge is stored in a knowledge base 7

Two Knowledge Types n Facts Procedures (usually rules) n Rules n IF-THEN-ELSE n Explanation Capability n 8

Rules (Example) IF interest_rate > 10 AND loan >= 10, 000 THEN review_loan = True. n n premise conclusion If premise is true, the rule is said to be triggered. A rule fires implies that the action specified by the conclusion clauses is taken.

Three Major ES Components n n n Knowledge Base Inference Engine User Interface 10

Three Major ES Components User Interface Inference Engine Knowledge Base 11

Knowledge Base n n n The knowledge base contains the knowledge necessary for understanding, formulating, and solving problems Two Basic Knowledge Base Elements n Facts n Special heuristics, or rules that direct the use of knowledge Knowledge is the primary raw material of ES 12

Inference Engine n n n The brain of the ES The control structure (rule interpreter) Provides methodology for reasoning (inference techniques) 13

User Interface n n Language processor friendly, problem-oriented communication NLP (natural language processing), or menus and graphics 14

Explanation Subsystem (Justifier) n n Other ES component. Traces and explains the ES behavior by interactively answering questions: -Why? -How? -What? -(Where? When? Who? ) 15

The Knowledge Engineer n 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 Usually also the System Builder 16

Inference Techniques (1) n Deductive Reasoning n n Combines facts (axioms) with general knowledge in the form of implications to conclude new facts. Example: Axiom: I am sleeping in class Implication: Sleeping in class => Rude shock in exams Conclusion: I will get a rude shock in the exam. Modus Ponens: IF A is true, and if A => B, then B is true.

Inference: Forward Chaining n n Data driven reasoning. Attempts to conclude a hypothesis (goal) from available information. Reasoning starts from available data and continues from that data, and when it was sufficient, conclusion will be drawn. Useful to find a hypothesis (goal) from a limited information (planning, control, interpreter etc)

Forward Chaining: Example R 1: R 2: R 3: R 4: R 5: n IF IF IF A and C THEN E D and C THEN F B and E THEN F B THEN C F THEN G Given facts: A is true B is true What can be concluded? Cycle through rules, looking for rules whose premise matches the working memory. Working memory A, B R 4 fires: assert new fact C A, B, C R 1 fires: assert new fact E A, B, C, E R 3 fires: assert new fact F A, B, C, E, F R 5 fires: assert new fact G A, B, C, E, F, G

Inference Techniques (2) n Inductive Reasoning n Generalizing from specific facts: case 1: Game on 21 st Sept. (Friday), It rained, We lost. case 2: Game on 5 th Nov. (Friday), It rained, We lost. …. . Induce general rule: IF game is on a Friday AND it rains Then we lose.

Inference: Backward chaining n n Goal driven reasoning. Attempts to prove a hypothesis (goal) by gathering supporting information. Reasoning starts from potential goal and from there, try to find any evidence (facts) to support/reject the goal. Useful when goal was known earlier, just need to find information to support the goal (diagnostic, debugging etc)

Backward chaining: Example Goal: I What data to support? Rules R 1: IF R 2: IF R 3: IF R 4: IF R 5: IF Goal I: need to trigger R 2 B and C THEN G A and G THEN I D and G THEN J E or F THEN C D and C THEN K Need both A and G Subgoal A: need user input (ask user) Subgoal G: Need to trigger R 1 Need both B and C Subgoal B: need user input (ask user) Subgoal C: Need to trigger R 4 Need E or F Subgoal E: need user input (ask user)

Problem Areas Addressed by Expert Systems n n n n Prediction systems Diagnostic systems Design systems Planning systems Monitoring systems Debugging systems Repair systems Control systems 23

Expert Systems Benefits n n n Increased output and productivity Decreased decision making time Increased process(es) and product quality Capture scarce expertise Accessibility to knowledge and help desks 24

Problems and Limitations of Expert Systems Knowledge is not always readily available n Expertise can be hard to extract from humans n Each expert’s approach may be different, yet correct n Experts’ vocabulary often limited and highly technical n 25

Expert Systems and the Web/Internet/Intranets n n Provide knowledge and advice Help desks Spread of multimedia-based expert systems (Intelimedia systems) Support ES and other AI technologies provided to the Internet/Intranet 26

ES Shell n n Includes All Generic ES Components But No Knowledge n EMYCIN from MYCIN n (E=Empty) 27

Expert Systems Shells Software Development Packages n Exsys n n K-Vision n n http: //www. exsys. com http: //www. ginesys. com/kvision/index. htm e. Xpertise 2 Go n http: //www. expertise 2 go. com/webesie/ 28

An Expert System at Work Exsys Demo http: //www. exsys. com/demomain. html 29

Intelligence Density Dimension n n Explainability Flexibility Scalability Independence from experts