2. What is Expert System?
- Intelligent program that solves problem in a narrow domain area
- Make use of specific expert knowledge rather than algorithm
- Simulates the decision making process of a human expert in a specific
domain
- Domain refers to the area within which the task is being performed
- Tends to substitute the human expert
- Example : MYCIN (An expert system for treating blood infections. It
diagnoses patients based on reported symptoms and test results. It
recommends a course of treatment. It uses backward chaining for
reasoning.)
3. Comparison : Human Expert Vs. Expert System
Factors Human Expert Expert System
Life Perishable Permanent
Processing Slow Fast
Performance Unpredictable Consistent
Replication Slow Quick
Cost Expensive Affordable
Scope Broad focus Narrow focus
Creative Ability Present Absent
Growth Adaptive Instructional
Base Common Sense Knowledge Base
4. Necessity of Expert System
1. Enable the use of expertise at remote locations where human experts
refuse to go or work
2. Automate routine tasks that requires human expertise all the time
3. Replace a human expert due to various conditions
4. Reduce operational cost
5. Help human expert in reducing the operational time, thus improving
productivity
5. Features of Expert System
1. Reasoning capacity
2. Deal with uncertainty
3. Use knowledge rather than data
4. Symbolic knowledge representation
5. Reliable problem solving
6. Advantages of Expert System
1. Provides consistent answer for repetitive problem
2. Maintains significant level of information
3. Proper explanation of decision making
4. Performance does not depends upon the emotions and mood
7. Disadvantages of Expert System
1. Lack of common sense
2. Unable to make creative response in unusual situation
3. Error in knowledge base leads to wrong decision
4. Unable to adapt to changing environment
9. 1. Knowledge Base
- Data structure containing expert knowledge of the domain
- Knowledge represented in the form of rules, generally IF...THEN
- The accuracy of expert system depends on collection of accurate and
precise knowledge
- Knowledge may be factual or heuristic
Architecture of Expert System
10. 2. Working Memory
- Data structure containing problem specific knowledge
- It consists of the problem to be solved
Architecture of Expert System
11. 3. Inference Engine
- Set of procedures for matching knowledge base with problem specific
working memory
- Helps in deducting the problem to find the solution
- Efficient inference engine provides correct and flawless solution
- Makes use of rule based system with strategies like forward chaining or
backward chaining
Architecture of Expert System
12. 4. User Interface
- Provides interaction between user and the expert system
- User may not be expert
- It makes it easy to trace the credibility of the deductions
Architecture of Expert System
13. 1. Knowledge Acquisition
- It deals with how to acquire required domain knowledge
- Finding knowledge from the human experts
- Converting the acquired knowledge into rules
- Injecting the developed rules into knowledge base
Development of Expert System
14. 2. Knowledge Representation
- Involves representation of knowledge in computer acceptable form
- Representation may be logical or structured
- Logical (FOPL)
- Structured (Semantic net, frames)
Development of Expert System
15. 3. Knowledge Inferencing
- Involves acquiring new knowledge from the existing one
- Makes use of rule based reasoning (Forward chaining or Backward
chaining)
Development of Expert System
16. 4. Knowledge Transfer
- Involves organizing, capturing and distributing knowledge to ensure
availability for future users
Development of Expert System