Google
 

Friday, December 21, 2007

The Human Element in Expert Systems


The Human Element in Expert Systems
At least two humans, and possibly more participate in the development
and use of an expert system.
 The Expert
The expert is a person who has special knowledge, judgment,
experience, and methods along with the ability to apply these talents to
give advice and solve problems.
 The knowledge engineer
The knowledge engineer helps the expert structure the problem area
by interpreting and integrating human answers to questions, drawing
analogies, posing counter examples, and bringing to light conceptual
difficulties.
 The user
Most computer based systems have evolved into a single user mode.
In contrast, an ES has several possible classes of users.
o A non expert client seeking direct advice. In such case, the ES
act as a consultant or advisor.
o A student who wants to learn. The ES acts as an instructor.
o An ES builder who wants to improve or increase the knowledge
base. Here the ES acts as a partner.
 How Expert system work
ES construction and use consists of three major activities.
1. Development
2. Consultation
3. Improvement
 Development
The development of an expert system involves the construction of the
knowledge base by acquiring knowledge from experts or documented
sources.
Development activity also includes the construction (or acquisition) of
an inference engine, a blackboard, an explanation facility, and any
other required software, such as interfaces.
The major participants in this activity are the domain expert, the
knowledge engineer, and possibly information system analysts and
programmers.
A tool that is often used to expedite development is called the ES shell.
ES shells include all the generic components of an ES, but they do not
include the knowledge.
 Consultation
Once the system is developed and validated, it is deployed to the
users, who are typically novices.
When users seed advices, they access the ES.
 Improvement
ES are improved several times through a process called rapid
prototyping during their development.
 Problem Areas addressed by Expert System
Prediction
Systems
This includes weather forecasting, demographic predictions, economic
forecasting, traffic predictions, crop estimates, and military, marketing, or
financial forecasting.
Diagnostic This system typically relates observed behavioral irregularities to underlying
Systems causes.
This includes medical, electronic, mechanical, & software diagnosis.
Design
Systems
This develops configurations of objects that satisfy the constraints of the
design problem.
Such problems include circuit layout, building design, and plan layout.
Planning

Systems
This specializes in problems of planning, such as automatic programming.
They also deal with short and long term planning areas such as project
management, routing, communications, product development, military
applications, and financial planning.
Monitoring
Systems
This compares observations of system behavior with standards than seem
crucial for successful goal attainment.
Air traffic control to fiscal management tasks
Debugging
Systems
This relies on planning, design, and prediction capabilities to create
specifications or recommendations for correcting a diagnosed problem.
Repair
Systems
This develops and executes plans to administer a remedy for certain
diagnosed problems.
Such systems incorporate debugging, planning and execution capabilities.
Instruction
Systems
This incorporates diagnosis and debugging subsystems that specifically
address the student needs
.
Control
Systems
These adaptively govern the overall behavior of a system.

No comments: