Expert System in artificial intelligence

All about expert systems in Artificial intelligence

1. 1
CHAPTER 10
Knowledge-Based Decision
Support: Artificial Intelligence and
Expert Systems

2. 2
Knowledge-Based Decision
Support: Artificial
Intelligence and Expert
Systems
Managerial Decision Makers are
Knowledge Workers
Use Knowledge in Decision Making
Accessibility to Knowledge Issue
Knowledge-Based Decision Support:
Applied Artificial Intelligence
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

3. 3
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

4. 4
Basic Concepts Of Expert
Systems
Expertise
Transferring Experts
Inferencing
Rules
Explanation Capability
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

5. 5
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

6. 6
Some Facts about
Expertise
Expertise is usually associated with a high
degree of intelligence, but not always with the
smartest person
Expertise is usually associated with a vast
quantity of knowledge
Experts learn from past successes and
mistakes
Expert knowledge is well-stored, organized and
retrievable quickly from an expert
Experts have excellent recall
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

7. 7
Experts
Degrees or levels of expertise
Nonexperts outnumber experts
often by 100 to 1
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

8. 8
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

9. 9
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

10. 10
Two Knowledge Types
Facts
Procedures (usually rules)
Regarding the Problem Domain
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ
•Deep knowledge
Knowledge base contains complex knowledge
•Self-knowledge
Able to examine own reasoning
Explain why conclusion reached

11. 11
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

12. 12
Rules
IF-THEN-ELSE
Explanation Capability
–By the justifier, or explanation
subsystem
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

13. 13
Explanation Capability
Explain advice or
recommendations
–By the justifier, or explanation
subsystem

14. 14
Applications

Finance
–Insurance evaluation, credit analysis, tax planning, financial
planning and reporting, performance evaluation
Data processing
–Systems planning, equipment maintenance, vendor evaluation,
network management

Marketing
–Customer-relationship management, market analysis, product
planning

Human resources
–HR planning, performance evaluation, scheduling, pension
management, legal advising
Manufacturing
–Production planning, quality management, product design, plant
site selection, equipment maintenance and repair

15. 15

16. 16
Structure of
Expert Systems
Development Environment
Consultation (Runtime)
Environment
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

17. 17

18. 18
Three Major ES
Components
Knowledge Base
Inference Engine
User Interface
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

19. 19
Three Major ES
Components
User
Interface
Inference
Engine
Knowledge
Base
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

20. 20
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
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

21. 21
Knowledge Acquisition
Subsystem
Knowledge acquisition is the
accumulation, transfer and
transformation of problem-solving
expertise from experts and/or
documented knowledge sources to a
computer program for constructing
or expanding the knowledge base
Requires a knowledge engineer
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

22. 22
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

23. 23
Inference Engine
The brain of the ES
The control structure (rule interpreter)
Provides methodology for reasoning
It provides direction about how to use
system’s knowledge by developing the
agenda that organizes and controls
the steps taken to solve problems
whenever consultation takes place.
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

24. 24
User Interface
Language processor for friendly,
problem-oriented communication
menus and graphics
–Scrolling dialog interface: It is easiest to implement
and communicate with the user.
–Pop-up menus, windows, mice are more
advanced interfaces and powerful tools for
communicating with the user; they require graphics
support.
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

25. 25
Blackboard (Workplace)
Area of working memory to
–Describe the current problem
–Record Intermediate results
Records Intermediate Hypotheses and
Decisions
1. Plan(how to attack on problem)
2. Agenda(Action execution)
3. Solution(candidate action)
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

26. 26
Explanation Subsystem
(Justifier)
Traces responsibility and explains the ES
behavior by interactively answering
questions
-Why?
-How?
-What?
-(Where? When? Who?)
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

27. 27
Knowledge refinement system
Knowledge Refining System
–Learning for improving performance
•Analyzes knowledge and use for learning and
improvements

28. 28
The Human Element in
Expert Systems
Expert
Knowledge Engineer
User
Others
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

29. 29
The Expert
Has the special knowledge,
judgment, experience and methods
to give advice and solve problems
Provides knowledge about task
performance
Define relationship among facts
which are important in solving
problem.
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

30. 30
The Knowledge Engineer
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
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

31. 31
The User
Possible Classes of Users
–A non-expert client seeking direct advice
(ES acts as a Consultant or Advisor)
–A student who wants to learn (Instructor)
–An ES builder improving or increasing
the knowledge base (Partner)
–An expert (Colleague or Assistant)
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

32. 32
Other Participants
System Builder
Systems Analyst
Tool Builder
Vendors
Support Staff
Network Expert
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

33. 33
How Expert Systems Work
Major Activities of
ES Construction and Use
Development
Consultation

Improvement
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

34. 34
ES Development
Knowledge base development
Knowledge separated into
–Declarative (factual) knowledge and
–Procedural knowledge
Development (or Acquisition) of an
inference engine, blackboard, explanation
facility, or any other software
Determine knowledge representations
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

35. 35
ES Shell
Includes All Generic ES
Components
But No Knowledge
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

36. 36
Consultation
Deploy ES to Users (Typically
Novices)
ES Must be Very Easy to Use
ES Improvement
–By Rapid Prototyping
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

37. 37
Improvement
ES Improvement
–By Rapid Prototyping
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

38. 38
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

39. 39

40. 40
Problem Areas Addressed
by Expert Systems
Interpretation systems
–Surveillance, image analysis, signal interpretation
Prediction systems
–Weather forecasting, traffic predictions, demographics
Diagnostic systems
–Medical, mechanical, electronic, software diagnosis
Design systems
–Circuit layouts, building design, plant layout
Planning systems
–Project management, routing, communications, financial
plans

41. 41
Problem Areas Addressed
by Expert Systems
Monitoring systems
–Air traffic control, fiscal management tasks
Debugging systems
–Mechanical and software
Repair systems
–Incorporate debugging, planning, and execution capabilities
Instruction systems
–Identify weaknesses in knowledge and appropriate remedies
Control systems
–Life support, artificial environment

42. 42
Expert Systems Benefits
Increased Output and Productivity
Decreased Decision Making Time
Increased Process(es) and Product Quality
Reduced Downtime
Capture Scarce Expertise
Flexibility
Easier Equipment Operation
Elimination of Expensive Equipment
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

43. 43
Operation in Hazardous Environments
Accessibility to Knowledge and Help Desks
Can Work with Incomplete or Uncertain Information
Provide Training
Enhancement of Problem Solving and Decision Making
Improved Decision Making Processes
Improved Decision Quality
Ability to Solve Complex Problems
Knowledge Transfer to Remote Locations
Enhancement of Other Information system
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

44. 44
Lead to
Improved decision making
Improved products and customer
service
May enhance organization’s image
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

45. 45
Problems and Limitations
of Expert Systems
Knowledge is not always readily available
Expertise can be hard to extract from humans
Each expert’s approach may be different, yet
correct
Hard, even for a highly skilled expert, to work
under time pressure
Expert system users have natural cognitive
limits
ES work well only in a narrow domain of
knowledge
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

46. 46
Most experts have no independent means to
validate their conclusions
Experts’ vocabulary often limited and highly
technical
Knowledge engineers are rare and expensive
Lack of trust by end-users
Knowledge transfer subject to a host of perceptual
and judgmental biases
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

47. 47
Types of Expert Systems
Rule-based Systems
–Knowledge represented by series of rules
Frame-based Systems
–Knowledge represented by frames
Hybrid Systems
–Several approaches are combined, usually rules and frames
Model-based Systems
–Models simulate structure and functions of systems
Off-the-shelf Systems
–Ready made packages for general use
Custom-made Systems
–Meet specific need
Real-time Systems
–Strict limits set on system response times

48. 48
Expert Systems and the
Web/Internet/Intranets
1. Use of ES on the Net
2. Support ES (and other AI methods)
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

49. 49
Using ES on the Web
Provide knowledge and advice
Help desks
Knowledge acquisition
Spread of multimedia-based expert
systems (Intelimedia systems)
Support ES and other AI technologies
provided to the Internet/Intranet
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ