Sie sind auf Seite 1von 21

EXPERT SYSTEM

EXPERT SYSTEM
.

An Expert system is known as Knowledge based system is a


computer program that contain some of the subject specific
knowledge of one or more human experts. It uses question
answers, logical reasoning to get the solution of a problem.

An expert system is an interactive computer based decision tool


that uses both
facts and heuristics to solve difficult problems
based on knowledge acquired from an expert.

An expert system operates as an interactive system that


respond to questions ,asks for clarification , makes
recommendations and generally aids the decision making
process

Architecture of EXPERT
SYSTEM

EXPERT SYSTEM

Levels of Expert System

Need of an Expert
System
WHY EXPERT
SYSTEM ??????????????
Expert systems are necessitated by the limitations associated
with conventional human decision making process
Humans get tired from physical and mental workload.
Humans lie , hide and die.
Humans are unable to retain large amount of data in
memory.
Humans have limited working memory.
Humans are slow in recalling information stored in memory.

Features of Expert
System

Multiple result oriented


Conclusion based
Question answered Based
Logical Reasoning
Specific Knowledge rather than
specific Technique

Respond to questions

Makes recommendations

Working of expert system

Asks for clarification

Logical Structure

What

is Inference Rule

An inference rule is a rule made up of If clause and then clause. It is used to


make a decision .If the condition is true, the loop will enter into If loop
otherwise if statement is false, the loop will enter into ELSE loop.
The Inference Engine accepts user Input queries and responses to questions
through user interface and uses the dynamic information and static
information (the rules and facts ) stored in the knowledge base.
The inferring process is carried out in three stages
MATCH
SELECT
EXECUTE

If then Else Clause


If she is beautiful
then
If She is Highly Educated
then
If she ready to settle in Abroad
then MATCH Found
else
she is beautiful, educated but not willing to settle abroad
so MATCH FOUND
end If
else
she is Beautiful but not Educated so,MATCH NOT FOUND
End If
else
she is not beautiful , MATCH NOT FOUND
End If

Output of Inference Rule

At each step of If else Clause, a


conclusion is drawn which leads to
formation of another conclusion till
the final goal is Reached

EXPERT SYSTEMS
Process Multiple
Values for any
problem Parameter

PROBLEM :
If a User is Hunting
for any Pizza
Outlet

Results will be:


Dominos
Pizza Hut
Slice of Italy
Sabberros
Pizza Corner
Narula

Features of Expert
System

Multiple result oriented


Conclusion based
Question answered Based
Logical Reasoning
Specific Knowledge rather than
specific Technique

Backward Chaining and


Forward Chaining

Forward chaining moves from


known state to goal state.
Forward Chaining starts with
the data available and uses the
interference rules to conclude
more data until a desired goal is
reached.
It is Data Driven Technology
An interface Engine using
forward chaining
searches the interface rules
until it finds one in which the if
clause is true.
After If clause is true, after that
then clause is concluded and
information is added to data.

Backward Chaining usually


moves from Goal state to Original
State.

Backward Chaining starts with a


list of goals and works backward
to see if there is data which will
allow to conclude any of these
goals.
It is Goal Driven Technology

An Interface engine using


backward chaining would
search the inference rules until it
finds one which has a then clause
that matches the desired goal

Forward Chaining
ENCOURAGED (student)->MOTIVATED (student)
MOTIVATED (student) &HEALTHY (student)->WORKHARD
(student)
WORKHARD (student) & HASGOAL (student) ->EXCELL (student)
EXCELL (student) -> SUCCEDD (student)

As in example : we can chain forward to show that when student is


encouraged , is healthy and has some goals, the student will
suceed.

BACKWARD CHAINING
Patient (NOT WELL)->Patient (BLOOD TEST)
Patient (Not well and INFECTION DETECTED)-> PATIENT(REPORTS
CONCLUDED)PATIENT(SWINEFLU DETECTED)

Thfis process of setting up sub goals to confirm a goal continues


until all the sub goals are eventually satisfied or fail. If satisfied ,
the backward chain is established thereby confirming the main
goal

END USER Interactive


Dialog

This system is leading the user through a set of questions to


achieve the solution.
There is an interactive dialog system between User .Answers are
usually given in the form of yes or No .Even sometimes the
answer is given in the form of 1 and 0. 1 indicates yes answer 0
indicates No.
Another major difference is that Expert systems usually accept
answers in form odif YES and NO whereas Traditional System
will accept the where questions and answers in form of Why,
where, whose , then etc.

APPLICATIONS

Location of Faults in computer and communication systems


Identifications of Chemical Compound structures and Chemical
compounds.
Planning experiments in biology, Chemistry and molecular
genetics.
Stock and bond portfolio selection and management system.
Estate planning for minimal taxation and other specified goals.
Numerous area of law including civil case evaluation, product
liability, assault and battery and general assistance in locating
different law precedents.
Numerous applications related to space planning and exploration
Forecasting Crop Damage
Different type of Medical Diagnosis (internal medicine,
pulmonary diseases,
infectious blood diseases and so on )

Examples

DENDRAL
The first expert system to be completed was DENDRAL, developed
STANFORD university in the late 1960s. This system was capable
of determining the structure of Chemical compounds given a
specifications of the compound constituent elements and mass
spectrometry data obtained from the samples of compound.

MYCIN
It is an expert system which diagnosis infection blood diseases
and determines a recommended list of therapies for the patient.

Applications

Medical Diagnosis
Equipment Repair
Investment Analysis
Vehicle Routing
Contract Bidding
Production Control and Training

Das könnte Ihnen auch gefallen