Fundamentals of Artificial

Intelligence
Module Code & Version #

Introduction To Expert Systems

 Topic & Structure of The Lesson

• Expert system
• Categories of expert system
• Components of expert system
• Expert System shell
• Characteristics of expert system
• Expert System Life Cycle

Module Code and Module Title Title of Slides Slide 2 of 21

 Learning Outcomes

At the end of this topic, You should be able to

• Explain the expert system.
• Provide examples of application of expert system.
• Explain the categories of expert system.
• Explain the components of expert system.
• Define expert system shell with examples.
• Explain the characteristics of expert system.
• Explain the expert system life cycle.
• List the benefits and limitation of expert system.

Module Code and Module Title Title of Slides Slide 3 of 21

 Key Terms You Must Be Able To
Use
If you have mastered this topic, you should be able to use the following
terms correctly in your assignments and exams:

• Expert System
• Domain expert
• Knowledge Engineer
• Domain

Module Code and Module Title Title of Slides Slide 4 of 21

 What is an Expert System?

• An Expert System is a system that uses human

knowledge to solve problems that would normally
be solved by human experts.

• Well-designed systems mimic the reasoning process

that experts use to solve problems in a specific
domain.

• It uses both facts and heuristics to solve complex

problems in specific domain.

Module Code and Module Title Title of Slides Slide 5 of 21

 Problem Domain vs Knowledge
Domain
• An expert’s knowledge is specific to
one problem domain – medicine,
finance, science, engineering, etc.

• The expert’s knowledge about solving

specific problems is called the Problem Domain

knowledge domain.

• The problem domain is always a Knowledge

Domain
superset of the knowledge domain.

Module Code and Module Title Title of Slides Slide 5 of 21

 Functions of an expert system

An expert system represents A simple expert system may

the knowledge of one or more represent the knowledge of a
domain experts. single expert.

The purpose of an expert The expert system can provide

system is to provide advice to a this consultation through a
user through a process of series of questions to the user.
consultation.
Module Code and Module Title Title of Slides Slide 6 of 21
 Quick Review Question

Identify the domain of expertise for which each of the following human
experts is well known
• Jamie Oliver
• Gary Lineker
• Patrick Moore
• Stephen Hawking

Module Code and Module Title Title of Slides Slide 8 of 21

 Applications of Expert System

• Expert systems have been developed for a wide

range of domains.
• These include systems giving medical, legal or
financial advice, but the range of domains is
increasing.

Module Code and Module Title Title of Slides Slide 10 of 21

 Applications of Expert System

Application Description
Design Domain Camera lens design, automobile design.
Diagnosis Systems to deduce cause of disease
Medical Domain from observed data, conduction medical
operations on humans.
Comparing data continuously with observed
Monitoring Systems system or with prescribed behavior such as
leakage monitoring in long petroleum pipeline.

Process Control Controlling a physical process based on

Systems monitoring.

Knowledge Domain Finding out faults in vehicles, computers.

Detection of possible fraud, suspicious

Finance/Commerce transactions, stock market trading, Airline
scheduling, cargo scheduling.

Module Code and Module Title Title of Slides Slide 10 of 21

 Applications of Expert System
Agriculture Domain
POMME, an expert system for apple orchard management.
COMAX, a crop management expert system for cotton.

Medical Domain
CaDet, an expert system that can classified tumor at initial
phase.
PXDES is used to analyze degree of lung cancer

Design Domain
FSSE (Feng Shui Expert Evaluator). This system will identify
whether a location fulfill the feng shui criteria and suggest to
the architect on the best angle to design the building.

Module Code and Module Title Title of Slides Slide 10 of 21

 Applications of Expert System
Automobile Domain
EXEDS (expert engine diagnosis system). It assists auto
mechanics in fault diagnosis of engines by providing systematic
and step-by-step analysis of failure symptoms and offering
maintenance or service advice.

Note:
Expert system can be implemented in other domains such as tourism, commercial,
education, etc.

Module Code and Module Title Title of Slides Slide 10 of 21

 Categories of Expert System

All expert systems produce advice as output but the way that
the output is used determines the category of the expert
system. There are several ways of categorizing the output
provided by an expert system.

The output provided by expert systems can be used for:

• Advice
• Classification
• Diagnosis
• Planning
Output

Module Code and Module Title Title of Slides Slide 13 of 21

 Categories of Expert System
Expert System use to for advice
If an expert system is used to give information to the user, then
the output from such an expert system is used to advise on a
course of action.

For example:
• Companies with no legal department use expert systems to
advise on certain aspects of the law.
• Oil companies use expert systems like GeoQuest and GeoPlay
to advise on the most likely places to drill for oil.
• British Gas use expert systems to predict where corrosion is
most likely to occur in the gas supply system.

Module Code and Module Title Title of Slides Slide 13 of 21

 Categories of Expert System
Expert systems used for classification
If an expert system is used to identify or grade information, then
the output from such an expert system is used to classify
information.

For example:
• Botanists use expert systems to classify rare plants
• Geologists use expert systems to categories minerals
• Thesys is an expert system that is used in some universities to
allow students to grade their own written reports before
submitting them
• Expert systems are used by many supermarket chains to
identify patterns of customer spending.

Module Code and Module Title Title of Slides Slide 13 of 21

 Categories of Expert System
Expert systems used for diagnosis
If an expert system is used to predict or determine the cause of a
problem, then the output from such an expert system is used to
help to diagnose problems.

For example:
• Medical expert systems are used to diagnose illnesses and
diseases.
• Government agencies are using expert systems for
troubleshooting and intelligent alternative selection.
• Airlines use expert systems to help the pilot to diagnose
problems with the aircraft.
• GEC Marconi used an expert system called APEX to identify faults
in complex circuit boards – in fact, diagnostic expert systems are
widely used to help speed up the repair of complex equipment.

Module Code and Module Title Title of Slides Slide 13 of 21

 Categories of Expert System
Expert systems used for planning
If an expert system is used to design or prepare an itinerary or
schedule, then the output from such an expert system is used to
help plan.

For example:
• The SUMit expert system was used by KLM airlines to plan
employee rotes.
• The expert system used by the Standard Communications
Company is used to plan the design of fibre optic cables
• The expert system used by Lufthansa aircraft is used to plan the
flight path for an aircraft’s journey
• Military expert systems help provide strategic advantage over
the enemy by assisting in battle assessment

Module Code and Module Title Title of Slides Slide 13 of 21

 Components of Expert System
Expert System Shell

Knowledge
Base

Human Inference
Knowledge
Expert Engine
Engineer

User
Interface

End User
(Maybe non-
expert)

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System

Human Expert
Someone who is an expert where his/her knowledge is taken to
develop an expert system.

Knowledge Engineer
Knowledge engineer is a technical person who collects and
integrates knowledge into the knowledge base.

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System
Knowledge Base
• Knowledge base is a repository of facts. It stores all the
knowledge about the problem domain. It is like a large
container of knowledge which is obtained from different
experts of a specific field.

• There two types of knowledge store in the knowledge base.

• Factual knowledge - It is the information widely accepted
by the knowledge engineers and scholars in the task
domain.
• Heuristic knowledge - It is about practice, accurate
judgement, one’s ability of evaluation, and guessing.

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System
• The knowledge in the knowledge base can be represented in
rules based (production rules) or frame – this will cover in
other chapter.

Database vs. Knowledge base

• Database is a collection of data representing facts in their
basic form.
• Knowledge base stores information such as answers to
questions or solutions to problems.

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System
Inference Engine
• Inference engine interprets and evaluates the facts in the
knowledge base in order to provide an answer.

• In case of knowledge-based ES, the inference engine acquires

and manipulates the knowledge from the knowledge base to
arrive at a particular solution.

• In case of rule based ES, the inference engine controls the

order in which production rules are applied and resolves
conflicts if more than one rule is applicable at a given time.

• To recommend a solution, the Inference Engine uses the

following strategies:
- Forward Chaining
- Backward Chaining
Module Code and Module Title Title of Slides Slide 15 of 21
 Components of Expert System

Forward Chaining
• It is a strategy of an expert system to answer the
question, “What can happen next?”
• Here, the inference engine follows the chain of conditions
and derivations and finally deduces the outcome. It considers
all the facts and rules, and sorts them before concluding to a
solution.

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System

Forward Chaining
• This strategy is followed for working on conclusion, result, or
effect. For example, prediction of share market status as an
effect of changes in interest rates.

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System
Backward Chaining
• With this strategy, an expert system finds out the answer to
the question, “Why this happened?”.
• On the basis of what has already happened, the inference
engine tries to find out which conditions could have
happened in the past for this result.

• This strategy is followed for finding out cause or reason. For

example, diagnosis of blood cancer in humans.

Module Code and Module Title Title of Slides Slide 15 of 21

 Components of Expert System

User Interface
• User interface provides interaction between user of the ES
and the ES itself. It is generally a natural language processing,
takes the user's query in a readable form and passes it to the
inference engine.
• After that, it displays the results to the user. In other words,
it's an interface that helps the user communicate with the
expert system.

Module Code and Module Title Title of Slides Slide 15 of 21

 Expert System Shell

• It provide framework that is use for development of an

expert system.

• It consists of an inference engine and user interface but

without the knowledge base.

• This allows the non-programmers to add their own

knowledge on a problem of similar structure but to re-use
the reasoning mechanism.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Shell

• One shell can be use in many different domains.

• It is very useful if the match between the problem and the

shell is good, but they are inflexible.

• It work best in diagnostic and advice-style expert system.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Shell

Module Code and Module Title Title of Slides Slide 16 of 21

 Characteristics of Expert System
High level Performance
• The system must be capable of responding at a level of
competency equal to or better than an expert system in the
field.
Domain Specificity
• Expert systems are typically very domain specific.
• The developer of such a system must limit his or her scope of
the system to just what is needed to solve the target
problem.
Good Reliability
• The expert system must be as reliable as a human expert.
Understandable
• The system should be understandable i.e. be able to explain
the steps of reasoning while executing.

Module Code and Module Title Title of Slides Slide 4 of 21

 Characteristics of Expert System

Use symbolic representations

• Expert system use symbolic representations for knowledge
(rules, networks or frames).
• Perform their inference through symbolic computations that
closely resemble manipulations of natural language.
Adequate Response time
• The system should be designed in such a way that it is able to
perform within a small amount of time, comparable to or
better than the time taken by a human expert to reach at a
decision point.

Module Code and Module Title Title of Slides Slide 4 of 21

 Expert System Life Cycle

Knowledge Knowledge
Acquisition Representation

System
Validation

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Knowledge Acquisition

• It is process of getting information from the human expert

or from the chosen source and into form required by the
expert system.
• This will be done by the knowledge engineer – person who
will be designing & developing the expert system.
• There are FIVE (5) techniques that can be use for knowledge
acquisition:
- Unstructured interview
- Structured interview
- Focused discussion
- Role reversal
- Think-aloud

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Unstructured Interview

• It is open and exploratory – no fixed questions are prepared

and the interviewee is allowed to cover topics as he/she sees
fit.
• It can be used to set the scene and gather contextual
information at the start of the knowledge acquisition
process.
• There are THREE (3) types of question that can be use such
as probes, prompts and seed questions.
1. Seed question – helpful in starting an unstructured
interview – general question to start the conversation.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Unstructured Interview
• There are THREE (3) types of question that can be use such
as probes, prompts and seed questions.
2. A probe question – encourage the expert to provide
further information without indicating what that
information should be.
3. A prompts question – more directed and can help
return the interview to a relevant topic that is
incomplete.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Structured Interview
• A framework for the interview is determined in advance.
• It involves the use of check-lists and questionnaires to
ensure focus is maintained.
• Strict structure interview, allow the engineer to compare
answers between experts, less strict (semi-structured
interview) combine a focus detail with some freedom to
explore areas of interest.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Focused Discussion
• It is based on a particular problem or scenario – case study,
critical incident or specific solution.
• Case analysis – consider a case study that might occur in the
domain. The expert explains how it would be or was solve
by verbally or demonstration.
• Critical incident analysis – look at the unusual and probably
serious incidents.
• Critiquing – expert is asked to comment on someone else’s
solution to a problem or design.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Role reversal
• It switches the role between an engineer and expert.
• There are TWO (2) types of role reversal:
1. Teach-back interview – the engineer “teaches” the
expert on a subject that has already been discussed
and allows the expert to amend the knowledge if
necessary.
2. Twenty questions – engineer chooses a topic from a
predetermined set and the expert asks questions
about the topic in order to determine which one has
been selected.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Think-aloud
• It is use to elicit information about specific tasks.
• The expert is asked to think-aloud while carrying out the
task.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

Knowledge Representation

• It is where the knowledge is put into expert system.

• The knowledge engineer needs to decide on an appropriate
knowledge representation scheme; first order logic (predicate
logic), procedural-based (rules based), network-based (semantic
network), and frame-based.
• He/she also needs to decide on appropriate reasoning and search
strategies.
• At this point, the engineer is able to begin with prototyping the
expert system by using expert system shell or high level AI
language.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

High Level Programming Languages

• They are designed for AI, provide fast, flexible mechanism

for developing an expert system.
• Examples:
• PROLOG – support logic
• LISP – functional programming language
• OPS5 – production system language
• It demands certain programming skills from the user. So,
they are less suitable for a non-programmer.

Module Code and Module Title Title of Slides Slide 6 of 21

 Expert System Life Cycle

System Validation

• Perform the standard testing levels such as unit testing,

integration testing, system testing and acceptance testing.
• Unit testing – Test the individual components of an expert
system.
• Integration testing – Combine these components to test for
its connection between those components.
• System testing – Test the system as a whole.
• Acceptance testing – A test that will participate by the user
to provide feedback on the expert system.

Module Code and Module Title Title of Slides Slide 6 of 21

 Constructing Expert System

Identify the Identify the

Gathering domain
purpose of domain of the
knowledge.
developing the ES. expert system.

Testing the Implementing the Representing

system. system. domain knowledge

Module Code and Module Title Title of Slides Slide 9 of 21

 Benefits of Expert Systems

• Availability − They are easily available due to mass production of

software.
• Speed − They offer great speed. They reduce the amount of work an
individual puts in.
• Less Error Rate − Error rate is low as compared to human errors.
• Reducing Risk − They can work in the environment dangerous to
humans.
• Steady response − They work steadily without getting motional, tensed
or fatigued.

Module Code and Module Title Title of Slides Slide 11 of 21

 Limitations of expert systems

• Problem of knowledge acquisition – expert generally poor at

expressing their knowledge and knowledge engineer may not know
what they are looking for.
• Verification of the knowledge stored – knowledge may be internally
consistent but inaccurate, due to either expert error or
misunderstanding during the acquisition.
• Highly domain dependent and brittle – cannot fall back on the general
common sense or generalize their knowledge on unexpected cases.
• The user may not know the limitation of the system – user may not
know whether the conclusion is in fact false or knowledge base in
incomplete.

Module Code and Module Title Title of Slides Slide 12 of 21

 Summary of Main Teaching Points

• Expert systems use a different approach; they collect the small

fragments of human know-how into a knowledge-base which is used to
reason through a problem, using the knowledge that is appropriate.
• A different problem, within the domain of the knowledge-base, can be
solved using the same program without reprogramming. The ability of
these systems to explain the reasoning process through back-traces
and to handle levels of confidence and uncertainty provides an
additional feature that conventional programming doesn’t handle

Module Code and Module Title Title of Slides Slide ‹46› of 21

 Question and Answer Session

Q&A

Module Code and Module Title Title of Slides Slide ‹47› of 21

 What we will cover next

• Knowledge Representation (Part 1)

Module Code and Module Title Title of Slides Slide 21 of 21