Expert System:
Definition: An expert is a set of programs that manipulates
some encoded knowledge to solve problems in a specialized domain that normally
required human expertise. Once a sufficient body of expert knowledge has been acquired, it
must be encoded in some form, loaded into KBS, then tested and refined
continually throughout the life of the system.
Characteristic:
·
ES differs from ordinary computer systems in
various ways:
o
ES uses knowledge rather than raw data to
control the solution process – much of the knowledge is used in heuristic in
nature rather than algorithmic.
o
Knowledge is encoded and maintained as an entity separate from the control program- this
permits the incremental addition and refinement of the knowledge without
recompilation of the control program.
o
ES is capable of explaining how a particular
conclusion was reached and why requested information is needed during a
confirmation.
o
ES uses symbolic representation for knowledge (
script, frame, rule of FOPL, semantic network) and perform their inference
through symbolic computation by manipulating natural language,
o
ES reason with meta-knowledge, ES is based on Neural
network architecture.
Background History:
1)
First expert system to be completed was DENDRAL
developed by Stanford University, in the late 1960s. this system was capable of
determining the structure of chemical compounds given a specification of the
compound constituent element and mass spectrometry data obtained from samples of
the compound. It uses heuristic knowledge and thereby reduced the search space.
During the test, DENDRAL discovered a number
of structures previously unknown to an expert chemist.
2)
MYCIN: It's an ES that diagnoses infectious
blood disease and determines a recommended list of therapies for the
patient- MYCIN`s performance now exceeds that of experienced doctors.
Application:
- Different types of medical Diagnosis
- Diagnosis of complex electromechanical systems
- Diagnosis of software development projects
- Forecasting crop damage
- Identification of chemical compound structures and chemical compounds
- Assessment of geologic from dip meter logs
- Design of VLSI
- And Numerous military applications.
Architecture diagram
- The most common form of architecture used in expert and other types of KBS is the production System. This type of system uses knowledge encoded in the form of production rules, that is if … then rule.
- If: condition1 & condition2 & condition
then: take action4. - Each rule represents a small chunk of knowledge relating to the given domain of expertise.
- Inference in the production system is accomplished by a process of chaining through the rules recursively until a conclusion is reached or failure occurs.
· The most common form of architecture used in expert and other types of KBS is the Production System also called rule-based system.
· In the human modeling world, the rules are called Production, and the rule-based systems are called production systems.
· Each rule representations a small chunk of knowledge relating to the given domain of expertise.
· A number of related rules collectively may correspond to a chain of inferences which leads from some initially known facts to some useful conclusions.
Comments
Post a Comment