Download presentation
Presentation is loading. Please wait.
1
1 5.0 Expert Systems Outline 5.1 Introduction 5.2 Rules for Knowledge Representation 5.3 Types of rules 5.4 Rule-based systems 5.5 Reasoning approaches in rule-based systems 5.6 Forward chaining (or data-driven) 5.7 Backward Chaining (or goal-driven) 5.8 Conflict resolution 5.9 Rule-based expert systems 5.10 Architecture of an expert system.
2
2 5.1 Introduction AI System Knowledge Base (KB) Inference Mechanism Facts about objects: properties and relations Rules,Procedures, Theories Set of procedures to examine the KB in an orderly manner to answer questions, solve problems, or make decisions within the KB domain
3
3 5.2.Rules for Knowledge Representation Rules are used as one way to represent knowledge. Rules are expressed in the following form: If then Example: If temperature is below 10 then weather is cold Then-pattern, Conclusion or Consequent If-pattern, Premise Or Antecedent
4
4 Relation rules : If Temperature below 5 Then Weather is cold Recommendation rules: if it is cold AND name is “ALI” Then tell ALI “Wear a coat” Directive rules: If you want to sleep then close the door 5.3.Types of rules
5
5 5.4. Rule based systems Rule –based systems or production-rules systems are computer systems that use rules to provide recommendations or diagnosis, or to solve a particular problem. A rule-based system consists of three components: –A database of rules –A database of facts –An interpreter, or inference engine Knowledge base Inputs to the system The part of the system that controls the process of driving conclusions
6
6 5.5. Reasoning approaches in rule-based systems There are two inference approaches in rule- based systems: 1.Forward chaining (or data-driven): Starting from a set of antecedents we use deduction to reach a conclusion. 2.Backward Chaining( or goal-driven): Starting from a conclusion we try to prove it by following a logical path backward from the conclusion to a set of antecedents that are in the database of facts. Note: A rule is said to be triggered if its antecedents are matched with some of the facts in the facts data base. And it is said to be fired if its conclusion part is added to the facts data base or its consequent action is performed.
7
7 5.6.Forward chaining (or data-driven) Assume the following rules Rule 1: IF A^B THEN C Rule 2: IF A THEN D Rule 3: IF C^D THEN E Rule 4: IF B^E^F THEN H Rule 5: IF A^E THEN G Rule 6: IF D^E^G THEN I Facts: Fact 1 A Fact 2 B Fact 3 F Rules fired Rules triggered Facts 11,2A,B,F 22A,B,C,F 33A,B,C,D,F 44,5A,B,C,D,E,F 55A,B,C,D,E,F,H STOP6A,B,C,D,E,F,H,G
8
8 5.7. Backward Chaining Matching rules Goals Facts 5GA,B,F 3E 1C,DA,B,F 2DA,B,C, F StopA,B,C,D,E,F,G
9
9 5.8 Conflict resolution In case of forward chaining, if more than one rule are triggered at the same time, we must specify a strategy to select one of them (conflict resolution). There are many strategies: 1.Specifying a priority level for each rule, then select the rule with the highest priority level. 2.Longest matching strategy:select the rule with maximum number of matched antecedents. 3.Most recently added: select the rule which is triggered with the most recently added facts. 4.Using the order of the rule as a strategy: select the rule that appears first in the knowledge base. Note: the rules for solving the conflict resolution problem are called meta rules. And the knowledge associated with this part of the system is called meta knowledge.
10
10 5.9 Rule-based Expert Systems An expert system is one designed to model the behavior of an expert in some field, such as medicine or geology. Rule-based expert systems are designed to be able to use the same rules that the expert would use to draw conclusions from a set of facts that are present to the system. Persons involved in an expert system are: 1.The End-user: is the one who has the need for the system. 2.The domain expert: is the one who gives how conclusions and actions are determined depending on conditions in some field such as medicine or engineering. 3.The knowledge engineer: is the one who designs the rules for the system depending on the knowledge from the domain expert.
11
11 5.10. Architecture of an Expert System Inference Engine Explanation System Knowledge Base Editor User interface Knowledge Base Fact Database User Expert system Shell
12
12 Referring to the shown architecture of an expert system, we can define the parts of an expert system as follows: 1.Knowledge base: contains the specific domain knowledge that is used by an expert to derive conclusions from facts. 2.The facts data base: contains the case specific data that is to be used in a particular case to derive a conclusion. 3.The inference engine: is the part of the system that uses the rules and facts to derive conclusions. 4.The knowledge base editor: allows the engineer or the expert to edit the information in the knowledge base. 5.The explanation part: provides information to the user about how the inference engine arrived at its conclusions. 6.The user interface: provides access to the inference engine, the explanation system, and the knowledge base editor. An Expert System shell is an expert system with neither knowledge base nor facts sections.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.