CHAPTER 2

• What is an Expert System (ES)?


  A computer system that emulates the decision-making ability of a human expert.
• How does an Expert System solve complex problems?


  By reasoning through bodies of knowledge represented mainly as if–then rules.
• What is knowledge?

  A theoretical or practical understanding of a subject or domain.
• What are individuals who possess knowledge called?

  Experts.
• How is knowledge categorized?

  Into tacit and explicit knowledge.
• What defines a knowledgeable and skilled person?

  They are capable of solving problems in a specific area or domain.
• What must an expert be able to do in the context of an expert system?


  Communicate their knowledge.
• What is the role of the expert in the development of an expert system?


  To participate in the expert system development and commit time to the project.
• Who is the most important player in the expert system development team?


  The expert.
• What is the role of a domain expert?


  Leader of the expert system development team.
• What are the primary roles of a domain expert?


  Ensure deliverables and milestones are met, and interact with the team.
• What is the role of a project manager in an expert system development team?


  A computer scientist who designs and implements programs incorporating AI techniques.
• What does a knowledge engineer do?


  Supplies knowledge and structures it into an expert system.
• What is the responsibility of a programmer in an expert system?


  To describe the domain knowledge in terms that a computer can understand.
• Who is considered an end-user?


  A person who uses the expert system when it is developed.
• What must an end-user feel about the expert system?


  Confident in its performance and comfortable using it.
• What is the structure of a rule-based Expert System?


  • Knowledge base: Contains domain knowledge as rules (IF-THEN).
  • Database: Set of facts for matching against rules.
  • Inference engine: Carries out reasoning to reach solutions.
  • Explanation facilities: Allows users to understand conclusions.
• What does the knowledge base contain?


  Domain knowledge useful for problem solving.
• How is knowledge represented in the knowledge base?


  As a set of rules with IF (condition) THEN (action) structure.
• What happens when the condition part of a rule is satisfied?


  The rule is said to fire and the action part is executed.
• What does the database include?


  A set of facts used to match against the IF parts of rules.
• What is the function of the inference engine?


  To carry out reasoning that leads to a solution.
• What do explanation facilities enable users to do?


  Ask how a conclusion is reached and why a specific fact is needed.
• How do expert systems compare to conventional systems and human experts?


  They can make mistakes like human experts.
• Can expert systems make mistakes?


  Yes, they can make mistakes just like human experts.
• What is a Decision Tree (DT)?


  A representation of rules in a knowledge base as connected decision points.
• What is the structure of a Decision Tree?


  It has a starting decision (root) and ongoing options chosen based on conditions.
• What happens at each leaf of a Decision Tree?


  An action is executed.
• Why is a decision tree considered efficient?


  Because decisions are simple, testing only one condition at a time.
• How can Boolean combinations of tests be represented in a Decision Tree?


  By using tree structures to represent AND/OR conditions.
• How does an Expert System reach a conclusion?


  By matching the rule IF parts to the facts, producing inference chains.
• What is the match-fire procedure in an inference engine?


  It cycles through matching rules to facts and firing applicable rules.
• What is forward chaining?


  A data-driven reasoning process starting from known data.
• What happens when a rule is fired in forward chaining?


  A new fact is added to the database.
• When does the match-fire cycle stop in forward chaining?


  When no further rules can be fired.
• What is a weakness of forward chaining?


  It may execute rules unrelated to the established goal.
• What is backward chaining?


  A goal-driven reasoning process that attempts to prove a goal.
• What does the inference engine do in backward chaining?


  It sets up a new goal to prove the IF part of a rule.
• How does backward chaining continue its process?


  By stacking rules until no more rules can prove the current sub-goal.
• When should forward chaining be chosen?


  When gathering information to infer from it.