expert system

10
Expert System Presented by: Mukul, Pankaj & Divya

Upload: mukul-singh

Post on 27-May-2015

918 views

Category:

Business


3 download

DESCRIPTION

Again classroom presentation March 2012.

TRANSCRIPT

  • 1. Expert SystemPresented by:Mukul, Pankaj & Divya

2. Expert System: Meaning Anexpert system is a computer systemthat competes with the decision-makingability of a human expert. Designed to solve complex problems byreasoning about knowledge, like anexpert, and not by following theprocedure of a developer as is the casein conventional programming. 3. Meaning Continued It is a computer program that represents andreasons with knowledge of some specialistsubject with a view to solving problems orgiving advice It is divided into two parts. Fixed, independent of the expert system: theinference engine. Variable: the knowledge base. Conversational: A dialog interface tocommunicate with users. 4. Software architecture Therule base or knowledge base In expert system technology, theknowledge base is expressed with naturallanguage rules IF ... THEN ... For examples :"IF it is living THEN it is mortal" "IF his age = known THEN his year of birth =date of today - his age in years. 5. The inference engine Theinference engine is a computerprogram designed to produce areasoning on rules. In order to produce a reasoning, it isbased on logic. With logic, the engine is able to generatenew information from the knowledgecontained in the rule base and data tobe processed. 6. Advantages Evolving an expert system is to add, modify ordelete rules. Since the rules are written in plainlanguage, it is easy to identify those to beremoved or modified. Quick availability and opportunity to programitself. The reliability of an expert system is the sameas the reliability of a database, i.e.good, higher than that of a classical program. 7. Disadvantages Knowledge collection and interpretation intorules, the knowledge engineering. Expert knowledge is not well understood:there is a lack of rules, rules arecontradictory, and some are poorly writtenand unusable. Expert systems most often use an engineincapable of reasoning. Therefore, oftentimesthe expert system works poorly and theproject is abandoned 8. On a lighter side 9. Application field Highly interactive or conversationalapplications, IVR, voice server. Fault diagnosis, medical diagnosis Decision support in complexsystems, process control, interactive userguide. Educational and tutorial software. Logic simulation of machines or systems Knowledge management. Constantly changing software. 10. Thank You IF= Nerds are bad, THEN = Politiciansare error !