JOMO KENYATTA UNIVERSITY

OF
AGRICULTURE & TECHNOLOGY
JKUAT SODeL

SCHOOL OF OPEN, DISTANCE AND eLEARNING

P.O. Box 62000, 00200
©2015

Nairobi, Kenya
E-mail: [email protected]

ICS 2405 Knowledge Based Systems

JJ II LAST REVISION ON January 14, 2015

J I
J DocDoc I
Back Close
 ICS 2405 Knowledge Based Systems
This presentation is intended to covered within one week.
The notes, examples and exercises should be supple-
mented with a good textbook. Most of the exercises have
solutions/answers appearing elsewhere and accessible by
JKUAT SODeL

clicking the green Exercise tag. To move back to the same

page click the same tag appearing at the end of the solu-
tion/answer.
©2015

Errors and omissions in these notes are entirely the re-

sponsibility of the author who should only be contacted
through the Department of Curricula & Delivery
(SODeL) and suggested corrections may be e-mailed to
[email protected].
JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 0
 ICS 2405 Knowledge Based Systems
LESSON 2
Understanding kbs

2.1. Problem Solving Techniques in AI

JKUAT SODeL

Recall, typical problem solving (and hence many AI) tasks can
be commonly reduced to:
• Modeling natural phenomena, and
©2015

• Search
Techniques that model natural phenomena generally concentrate
on the concept of knowledge representation
In a nutshell, some problems highlight search whilst others
knowledge representation
JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 1
 ICS 2405 Knowledge Based Systems
2.2. Knowledge Based Systems-Overview
Definition: a knowledge-based system is a software system ca-
pable of supporting explicit representation of knowledge and
of exploiting it through appropriate reasoning mechanism
JKUAT SODeL

in order to provide high-level problem solving performance.

• Knowledge based systems are systems that use knowledge
representation as their central concept.
©2015

• Knowledge based systems are developed for a specific task

domain
• It is therefore a specific, dedicated, computer based prob-
lem solver, able to face complex problems, which if solved
JJ II by man, would require advanced reasoning capabilities like:
J I – deduction,
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 2
 ICS 2405 Knowledge Based Systems
– hypothetical reasoning,
– model based reasoning,
– analogical reasoning,
• learning etc. Knowledge based systems are developed for
JKUAT SODeL

a specific task domain.

2.3. KBS-Tasks
©2015

Typical tasks for which knowledge based systems can be built

include:
• Diagnosis
• Scheduling

JJ II • Design
J I • Planning
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 3
 ICS 2405 Knowledge Based Systems
• Configuration

2.4. What is Knowledge?

• Knowledge is a theoretical or practical understanding of
JKUAT SODeL

a subject or a domain. Knowledge is also the sum of what

is currently known, and apparently knowledge is power.
• Those who possess knowledge are called experts.
©2015

• Anyone can be considered a domain expert if he or she

has deep knowledge (of both facts and rules) and strong
practical experience in a particular domain. The area of
the domain may be limited.

JJ II • In general, an expert is a skilful person who can do things

J I other people cannot.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 4
 ICS 2405 Knowledge Based Systems
2.4.1. Classification of Knowledge
• Criteria 1
• Priori Knowledge
– Universally true and cannot be denied without con-
JKUAT SODeL

tradiction e.g. mathematical laws, logical statements.

• Posteriori Knowledge
©2015

– Represents knowledge that is verified using sensory

experiences. Such knowledge can be denied based on
new knowledge without the need for contradictions.

• Criteria 2
• Procedural (Compiled) Knowledge
JJ II – Considers the manner in which things work under
J I different sets of circumstances. Includes step-by-step
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 5
 ICS 2405 Knowledge Based Systems
sequences and how-to types of instructions; it may
also include explanations.
• Declarative Knowledge
– Descriptive representation of knowledge. It tells us
JKUAT SODeL

facts: what things are e.g. “Elizabeth Kiiru is a young

Kenyan woman”.
• Tacit Knowledge
©2015

– Unconsciously knowing how to do something. By def-

inition, tacit knowledge is not easily shared. Tacit
knowledge consists often of habits and culture that
we do not recognize in ourselves. The concept of tacit
knowledge refers to a knowledge which is only known
JJ II to you and hard to share with someone else, which is
J I the opposite from the concept of explicit knowledge.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 6
 ICS 2405 Knowledge Based Systems
• Meta Knowledge
– Knowledge about knowledge.

2.4.2. Conceptual structure of a kbs (conceptual schema)

JKUAT SODeL

A knowledge-based system is generally composed of two parts:

• Central Part (Kernel)
©2015

– This implements the basic problem solving capabili-

ties of the knowledge-based system.
– The kernel is in turn composed of three main compo-
nents:
∗ Knowledge base (KB)
∗ Reasoning mechanism-inference engine.
JJ II
∗ Working memory
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 7
 ICS 2405 Knowledge Based Systems
• Peripheral Part This provides additional functions nec-
essary for a practical and effective use of the knowledge
based system e.g. user interface, explanation system, learner,
knowledge acquisition facility, etc.
JKUAT SODeL
©2015

JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 8
 ICS 2405 Knowledge Based Systems
JKUAT SODeL
©2015

Diagram-Conceptual Schema

2.4.3. Components of A Knowledge Based System

JJ II Based on the conceptual scheme above, a typical knowledge
J I based system has the following components.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 9
 ICS 2405 Knowledge Based Systems
• A knowledge base
• An inference engine
• A user interface
• Explanation facilities
JKUAT SODeL

• Learning facilities
The knowledge base contains the domain knowledge useful for
©2015

problem solving. In a rule-based expert system, the knowledge

is represented as a set of rules. Each rule specifies a relation,
recommendation, directive, strategy or heuristic and has the IF
(condition) THEN (action) structure. When the condition part
of a rule is satisfied, the rule is said to fire and the action part
JJ II is executed.
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 10
 ICS 2405 Knowledge Based Systems
•.1. The database/working memory includes a set of facts
used to match against the IF (condition) parts of rules stored in
the knowledge base.
JKUAT SODeL

The inference engine carries out the reasoning whereby the

KBS reaches a solution. It links the rules given in the knowledge
base with the facts provided in the database.
©2015

The explanation facilities enable the user to ask the KBS

how a particular conclusion is reached and why a specific fact is
needed. A KBS (specifically an expert system) must be able to
explain its reasoning and justify its advice, analysis or conclu-
JJ II sion.
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 11
 ICS 2405 Knowledge Based Systems
•.2. The user interface is the means of communication be-
tween a user seeking a solution to the problem and a KBS.
JKUAT SODeL
©2015

JJ II
J I •.3. Diagram Depicting components of KBS
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 12
 ICS 2405 Knowledge Based Systems
2.4.4. Criteria for choosing to implement a KBS
•.4. Only certain types of applications are suited to knowledge
based implementation. The following criterion is used in identi-
fying suitable domains. The fields under study:
JKUAT SODeL

• Should be able to be reduced to a series of rules rather then

mathematical formula or equations. – Knowledge based
systems not applicable if domain has a large number of
©2015

complex calculations.
• Is well understood so that well-defined knowledge can be
formulated and represented in Computer
• Should not encompass problems which take too short (i.e.
JJ II less than half an hour) or too long (larger than say one
J I week) a time to solve.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 13
 ICS 2405 Knowledge Based Systems
• Should have general agreement among recognized experts
in the domain to be computerized.
• The knowledge within the problem domain should be suf-
ficiently large to warrant the development of a knowledge-
JKUAT SODeL

based system.
• ‘Tame’ experts who are agreeable to involvement.
However, any application that requires access to specialist knowl-
©2015

edge is a potential area for knowledge based systems technology.

2.4.5. Characteristics of a KBSs

•.5. High-quality performance.

JJ II •.6. No matter how fast the system can solve a problem, the
J I user will not be satisfied if the result is wrong!
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 14
 ICS 2405 Knowledge Based Systems
•.7. On the other hand, the speed of reaching a solution is
very important. Even the most accurate decision or diagnosis
may not be useful if it is too late to apply, for instance, in
an emergency, when a patient dies or a nuclear power plant
JKUAT SODeL

explodes.

•.8. KBSs apply heuristics to guide the reasoning and thus

©2015

reduce the search area for a solution.

•.9. A unique feature of a KBS is its explanation capability.

It enables the system to review its own reasoning and explain
its decisions.
JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 15
 ICS 2405 Knowledge Based Systems
•.10. KBSs employ symbolic reasoning when solving a prob-
lem. Symbols are used to represent different types of knowledge
such as facts, concepts and rules.

•.11. We should be aware that an expert is only a human and

JKUAT SODeL

thus can make mistakes, and therefore, a KBS built to perform

at a human expert level also should be ”allowed” to make mis-
©2015

takes.

•.12. In KBSs, knowledge is separated from its processing (knowl-

edge base and inference engine are split up). A conventional
program is a mixture of knowledge and the control structure to
JJ II process this knowledge.
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 16
 ICS 2405 Knowledge Based Systems
•.13. When an expert system shell is used, a knowledge en-
gineer or an expert simply enters rules in the knowledge base.
Each new rule adds some new knowledge and makes the KBS
smarter.
JKUAT SODeL
©2015

JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 17
 ICS 2405 Knowledge Based Systems
2.4.6. KBS Comparison
JKUAT SODeL
©2015

2.5. Knowledge engineering

JJ II This refers to the process of developing knowledge-based systems.

J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 18
 ICS 2405 Knowledge Based Systems
•.14. Practitioners of Knowledge Engineering are called Knowl-
edge Engineers.

•.15. Thus the development of a Knowledge based system should

JKUAT SODeL

follow a professional approach with respect to:

• qualifications,
• personality, and
©2015

• attributes.

2.5.1. Process of Knowledge Engineering

Consists of five activities:
• Knowledge acquisition
JJ II
J I • Knowledge Validation
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 19
 ICS 2405 Knowledge Based Systems
• Knowledge Representation
• Inferencing and
• Explanation.
JKUAT SODeL

2.5.2. Tasks of knowledge engineers

• Extracting knowledge from people. (KA)
• Representing knowledge in some form. (KR)
©2015

• Including knowledge in a computer program which makes

use of the knowledge.
• Validating the software system produced.

JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 20
 ICS 2405 Knowledge Based Systems
2.6. Types of knowledge based systems
•.16. In general, knowledge based systems are classified accord-
ing to the human behavior they attempt to mimic. On the basis
of this, we have:
JKUAT SODeL

•.17. Expert Systems: They model the higher order cogni-

tive functions of the human mind They are used to mimic the
©2015

decision making process of the human mind.

•.18. Neural Networks: They model the brain at the biolog-

ical level They are adept at pattern recognition and introduce
the concept of learning into computing.
JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 21
 ICS 2405 Knowledge Based Systems
•.19. Case Based Reasoning: Models the human ability to
learn from past experience They borrow from the legal system
where past cases are used as a basis for making decisions in the
present cases.
JKUAT SODeL

•.20. Fuzzy Logic systems: Provide mechanisms for han-

dling uncertain knowledge
©2015

2.7. Knowledge acquisition

•.21. Knowledge acquisition is the process of extracting knowl-
edge (facts, procedures, rules) from human experts, books, docu-
ments, sensors or computer files and converting it into a form
JJ II that can be stored and manipulated by the computer for purposes
J I of problem solving.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 22
 ICS 2405 Knowledge Based Systems
•.22. It occurs throughout the entire development process for
a KBS.

2.7.1. Acquiring Knowledge

•.23. The important characteristics of knowledge are that it is
JKUAT SODeL

experiential, descriptive, qualitative, largely undocumented and

constantly changing.
©2015

•.24. There are certain domains where all these properties are
found and some where there are only a few.

•.25. The lack of documentation and the fact that experts carry
JJ II a lot of information in their heads, makes it difficult to gain
J I access to their knowledge for developing information systems in
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 23
 ICS 2405 Knowledge Based Systems
general and expert systems in particular.

•.26. Therefore, knowledge engineers have devised specialized

techniques to extract and document this information in an effi-
JKUAT SODeL

cient and expedient manner.

©2015

JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 24
 ICS 2405 Knowledge Based Systems
2.7.2. Knowledge Acquisition Process
JKUAT SODeL
©2015

JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 25
 ICS 2405 Knowledge Based Systems
•.27. Identification: Identify the problem including data, cri-
teria for solutions to meet, available resources, etc.

•.28. Conceptualization: Determine the key concepts and

JKUAT SODeL

relationships by characterizing the data, flow of information, the

domain structure, etc.
©2015

•.29. Formalization: Understand the underlying search space,

uncertainty issues, etc.

•.30. Implementation: Translate acquired knowledge into the

program.

JJ II
•.31. Testing: Validate and verify.
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 26
 ICS 2405 Knowledge Based Systems
2.7.3. Knowledge elicitation
•.32. It implies that knowledge acquisition is accomplished from
a human expert.
JKUAT SODeL

•.33. Interaction between the expert and a program or knowl-

edge engineer where the purpose of the program is to:
• Elicit knowledge in a systematic way,
©2015

• Store such knowledge in some representation, and

• Compile the represented knowledge into a runnable pro-
gram.

• Sources of Knowledge
JJ II • The Knowledge Engineer acquires knowledge from
J I a number of sources:
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 27
 ICS 2405 Knowledge Based Systems
• Written sources, such as books, manuals, standard procedures-
An important source of knowledge at the start, when the
knowledge engineer learns the vocabularies and general
topics of the domain.
JKUAT SODeL

• Experts
• Observations of the actual process
• Expertise of a procedural nature, in which a sequence of
©2015

tasks is performed, direct observation of the operation also

can provide input to the knowledge base.
• Experts make decisions based on an abstract and intellec-
tual process that even experts themselves have difficulty
JJ II understanding and verbalizing. The knowledge engineer
J I should elicit knowledge from the expert by helping the ex-
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 28
 ICS 2405 Knowledge Based Systems
pert to discover and communicate this process, using a
host of knowledge acquisition methods.

• Methods of Knowledge Elicitation

• Face to face interview with experts – knowledge engineers
JKUAT SODeL

interview the experts.

• Protocol analysis – This is a documentation of how the
©2015

expert behaves and processes information during problem

solving. Usually the experts think aloud. Observation –
The experts are observed at work.
• Questionnaires –This is where questions are sent to experts
for responses.
JJ II • Analysis of documented knowledge – It involves extraction
J I of knowledge from sources such as books, journals, articles,
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 29
 ICS 2405 Knowledge Based Systems
magazines, mass media materials.
• Rule induction (computer aided knowledge acquisition) –
rule induction can be viewed as a system that accepts ex-
amples and develops classification rules.
JKUAT SODeL

2.7.4. Issues with Knowledge Acquisition (Problems)

• Machine representation is lower in form than human usage
©2015

of knowledge;
• Many participants are involved and they have varied back-
grounds causing communication challenges
• Experts may not express their knowledge.

JJ II • Mismatch between the way experts hold their knowledge

J I and the way computers represent knowledge.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 30
 ICS 2405 Knowledge Based Systems
2.7.5. Characteristics of Knowledge Acquisition
• Knowledge acquisition is a labour and time intensive pro-
cess.
• Currently knowledge bases for knowledge based systems
JKUAT SODeL

are crafted by hand, this is a severe limitation on the rapid

deployment of such systems.
• Biggest ‘bottleneck’ in system development.
©2015

• Most expensive part (money, time & labour).

• Automating KA the ultimate goal.

2.8. Players in the Development Team

JJ II There are five members of the KBS development team:
J I • the domain expert
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 31
 ICS 2405 Knowledge Based Systems
• the knowledge engineer
• the programmer
• the project manager
• the end-user.
JKUAT SODeL

The success of their KBS entirely depends on how well the mem-
©2015

bers work together.

2.8.1. Domain Expert

• The domain expert is a knowledgeable and skilled person
capable of solving problems in a specific area or domain.

JJ II • This person has the greatest expertise in a given domain.

J I • This expertise is to be captured in the KBS.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 32
 ICS 2405 Knowledge Based Systems
• Therefore, the expert must:
– be able to communicate his or her knowledge
– be willing to participate in the expert system devel-
opment
JKUAT SODeL

– commit a substantial amount of time to the project.

• The domain expert is the most important player in the
KBS development team.
©2015

2.8.2. Knowledge Engineer

•.1. The knowledge engineer is someone who is capable of de-
signing, building and testing a KBS.

JJ II •.2. The knowledge engineer’s main tasks are:

J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 33
 ICS 2405 Knowledge Based Systems
• interviews the domain expert to find out how a particular
problem is solved.
• establishes what reasoning methods the expert uses to han-
dle facts and rules and decides how to represent them in
JKUAT SODeL

the expert system.

• chooses some development software or an expert system
shell, or looks at programming languages for encoding the
©2015

knowledge.
• responsible for testing, revising and integrating the expert
system into the workplace.

2.8.3. Programmer
JJ II •.3. The programmer is the person responsible for the actual
J I programming, describing the domain knowledge in terms that a
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 34
 ICS 2405 Knowledge Based Systems
computer can understand.

•.4. The programmer needs to have skills in symbolic program-

ming in such AI languages as LISP, Prolog and OPS5 and also
JKUAT SODeL

some experience in the application of different types of expert

system shells.
©2015

•.5. In addition, the programmer should know conventional

programming languages like Java, C, Pascal, FORTRAN and
Basic.

2.8.4. Project Manager

JJ II •.6. The project manager is the leader of the KBS development

J I team, responsible for keeping the project on track.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 35
 ICS 2405 Knowledge Based Systems
•.7. The project manager makes sure that all deliverable and
milestones are met, interacts with the expert, knowledge engi-
neer, programmer and end-user.

2.8.5. End-User
JKUAT SODeL

•.8. The end-user, often called just the user, is a person who
uses the KBS when it is developed.
©2015

•.9. The user must not only be confident in the KBS perfor-
mance but also feel comfortable using it.

•.10. Therefore, the design of the user interface of the KBS is

JJ II also vital for the project’s success; the end-user’s contribution
J I here can be crucial.
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 36
 ICS 2405 Knowledge Based Systems
Example . What is knowledge acquisition?
Solution:
Knowledge acquisition is the process of extracting knowledge
(facts, procedures, rules) from human experts, books, documents,
JKUAT SODeL

sensors or computer files and converting it into a form that can

be stored and manipulated by the computer for purposes of prob-
lem solving.
©2015

Exercise 1.  Describe the issues in knowledge acquisition. .

.

Learning Activities/assignments
JJ II
J I • Explain how you may acquire knowledge to build a system
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 37
 ICS 2405 Knowledge Based Systems
• Discuss ways of acquiring knowledge.
• Discuss problems associated with knowledge acquisition.
JKUAT SODeL
©2015

JJ II
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 38
 ICS 2405 Knowledge Based Systems
Solutions to Exercises
Exercise 1.

issues in knowledge acquisition

JKUAT SODeL

Machine representation is lower in form than human usage

of knowledge;
Many participants are involved and they have varied back-
©2015

grounds causing communication challenges(Keng., Dom. Ex-

pert, System designers, Users, etc.).
Experts may not express their knowledge);
Mismatch between the way experts hold their knowledge and
the way computers represent knowledge.
JJ II Exercise 1
J I
J DocDoc I
JKUAT: Setting trends in higher Education, Research and Innovation

Back Close 39