A2 Expert Systems

Expert system shell

Software used to create expert systems. User adds knowledge base and inference rules to create expert system in a particular field.

Factual Knowledge

The factual information acquired by the knowledge engineers from human experts.

User Interface

This allows a user to communicate with the system through either visual or aural prompts. Decisions will be output back to the user.

Inference Engine

This interrogates the knowledge base and draws inferences and conclusions based on the rules stored about the subject domain. It poses questions to the user and uses the answers provided by the user to determine a suitable response.

Heuristic knowledge

This is not simple logic/facts but an intuition developed through experience working in a particular field. Can be described as rules of thumb.

Fuzzy Logic

This uses probablitities/degrees of truth to make decisions. Decisions can be made when the information is incomplete, mimics human reasoning.

Benefits of Expert Systems

-Always available -Can support new/inexperienced staff -Makes rational decisions not emotional ones -Can make decisions faster than humans -Accurate -Cost effective -Consistent decision making process

Limitations of Expert Systems

-Doesn't learn from mistakes -Over reliance on technology -Not suitable for all problems -Cannot take account of special circumstances

Expert

A person with vast knowledge/skills in particular area. This may be a doctor, mechanic or insurance advisor. The knowledge engineer will collate information/rules from these by interviewing these.

Knowledge Engineer

A specialist who interviews and observes experts, verifies their findings and converts their words and actions into rules and facts in a knowledge base.

Knowledge Base

All of the rules/facts about a particular area. Developed by the knowledge engineer. Made from both factual and heuristic knowledge.

Expert System/ Knowledge based system

Computerised advisory programs that imitate the reasoning processes of experts in solving difficult problems. Consists of knowledge base, inference engine and user interface. Made using an expert system shell. Also known as knowledge based systems.

Binary Logic

Concepts which are either True or False.