KNOWLEDGE ACQUISITION

Knowledge acquisition is the process of acquiring, organizing, & studying knowledge.

Two main types of sources of knowledge –

 Documented (which can take many forms) and
 Undocumented (usually in the expert's mind).

There are two types of knowledge: shallow knowledge and deep knowledge.

Shallow knowledge: An expert has a base understanding of the subject, some of which could be described as
general. For example a pension’s expert will know that a pension is a source of income after retirement. Now,
obviously a great many non-experts know this. It is, however, central to the understanding of the pension’s
domain. Other general knowledge could be the general strategy that the expert employs when solving a
problem, the general class of problem that the expert is tackling and some typical problem solutions. This is
sometimes referred to as meta-knowledge.

Example: surface level information might be represented as: If the weather is bad then stay in bed.

Deep Knowledge: This is the knowledge that has been acquired by years of experience and study and is the
detailed `core` of the knowledge base. This knowledge covers such items as:
 Relationships between data
 Rules/concepts
 How important each item is
 What order he/she performs tasks
 The detailed strategies for arriving at solutions

The Knowledge Acquisition Process

i). Identification: - This is a stage in the problem is identified, its characteristics are examined and it is
broken down into smaller sub-problems. The participants are identified (experts, users, knowledge
engineers). The resources are identified or provided. The situation is studied.
ii). Conceptualization: -The best method of representing knowledge is identified and selected.
iii). Formalization: - The knowledge is extracted from the sources. Software and hardware issues are
resolved.
iv). Implementation: - The program is written or a prototype is developed.
v). Testing: - Test cases are used. The experts are consulted on acceptability of results.

Knowledge Elicitation: - It implies that knowledge acquisition is accomplished from a human expert.
Knowledge Acquisition Methods
Knowledge acquisition methods constitute tools used for the process of modeling knowledge. There are two
basic strategies in knowledge engineering:
 Starting from general and overall concepts, gradually leading the expert to elicit details of a topic.
 Starting from the details of specific cases and helping the expert establish and derive general
concepts from the specific examples.
Therefore, the methods of eliciting knowledge can be categorized into two broad groups:
i). Top-down (or deductive) methods
ii). Bottom-up (or inductive) methods

i). Top-Down Methods

The knowledge engineer organizes the acquisition sessions for discovering general concepts, rules, and
objects, and then gradually goes into the details of each concept, rule, or object.
This can be grouped into four categories:
1) Questioning methods
2) Object-oriented methods
3) Quantitative methods
4) Inventive methods

Knowledge Based Systems ~ Wainaina Page 1 of 9

1) Questioning Methods:
The knowledge engineer interviews the expert in a series of meetings, or asks the expert to fill out a
questionnaire.
Uses three strategies in questioning the expert
i). Structured interviews
 Designed to be systematic and goal-oriented.
 It forces a directed discussion on a topic; therefore it can reduce interpretation issues and may
also reduce the expert’s subjectivity.
ii). Unstructured interviews
 Informal: Good to use when beginning the KA process.
 They help the KE understand the domain and can help identify discussion points for structured
interviews.
iii). Questionnaires
 Used in cases where access to the expert is limited,
 A questionnaire is prepared for the expert to answer in writing.
 This method is useful for clarifying already developed topics in the advanced stages of
knowledge engineering.

2) Object - oriented Methods.

The KE focuses the interview sessions on discovering the objects within the domain. Usually done by asking
the expert to group the actual objects in the field in order to form a class of objects that has a common set of
attributes.

3) Quantitative methods.
The methods were developed in cognitive science and decision analysis for eliciting the degree of a decision
maker's preferences and utilities, and in grouping various objects and attributes
Used to measure and determine:
 The extent of relationships among objects (or concepts)
 The degree of uncertainty about the domain knowledge

4) Inventive methods.
Expert is allowed a more active part in the process in one of the following roles:
Expert as a teacher:
 Responsible for teaching and transferring expertise to the KE.
 Expert is given the responsibility for the preparation and organization of the elicitation sessions.
 Efficient at the early stages of knowledge acquisition
Expert as a partner in systematic innovation:
 This is an abstract concept
 Requires the expert and the KE to identify pieces of knowledge that are in contradiction, and to
discover solution methods for removing the contradiction
Example: The applicant's request for a loan and a less than favorable credit rating form a contradiction.
The expert must provide a solution for this contradiction, which leads to the elicitation of new pieces
of knowledge. The expert and knowledge engineer form a partnership in discovering contradictions
and creating solutions; this taps the expert's deep understanding of the domain.
Expert as the knowledge engineer
 The expert may have both technical interest in the system and the needed training in knowledge
engineering.
 The expert can also play the role of a knowledge engineer.

ii). Bottom-Up Methods

The knowledge engineer focuses the expert's attention on specific cases. This helps the expert abstract the
decision for resolving a specific case to a more generalized rule or concept.
The methods that can be grouped in this approach are
1) Example-based methods.

Knowledge Based Systems ~ Wainaina Page 2 of 9

2) Protocol analysis.
3) Observation of the expert's decision-making process.

1) Example-based methods
The example-based approach constitutes the foundation of case-based learning and learning by
analogy. In the example-based methods, the knowledge engineer and the expert work on a number of
representative cases or examples in one of the following ways:

a) Grouping examples: - The expert groups the examples based on their similarities and differences.
This process helps to determine categories of examples and the development of general rules for each
category.
b) Walk-through method: - The knowledge engineer selects a number of cases previously decided by
the expert, and asks the expert to walk through the decision process.
c) Quantitative analysis of examples: - The quantitative techniques are tools for helping the expert
discover the relations among various attributes of the decision cases.
Categorized into two groups:
i). Statistical methods
 The examples must be a random sample of the cases decided by the expert(s).
 The data on the examples are fed into a statistical technique, such as regression analysis, in
order to discover the expert's decision criteria.
ii). Inductive methods
 The example set contains a representative set of all possible cases the expert has encountered.
 The examples are fed into the inductive method, which produces a decision tree or a set of
decision rules.

2) Protocol analysis
The expert is asked to think aloud and verbalize his or her thought process while solving a set of actual
(or simulated) problems and making decisions. The KE records the process, and later analyzes the large
volume of information produced from this method to discover the general rules the expert uses in
solving problems.

3) Observation
It involves observation of the expert while solving a problem. Useful when the solution of the problem
is procedural and takes place in a sequence of steps through time
The absence of the biases and intrusion inherent in the KE’s questions makes this approach useful. The
KE must make sure that the expert is making decisions in the most realistic environment.

Computer Aided Knowledge Acquisition

Most KA techniques are VERY time consuming. The experts are often inarticulate. The knowledge engineers
are inexperienced and/or not sufficiently conversant with the domain. It is suggested that computers could be
used effectively in this process and this should:
 Increase productivity
 De-skill the KA process
 Eliminate the need for an expert

The use of computers for KA is rapidly expanding and this section considers a technique called rule induction
and mentions some of the tools available.

Automated Rule Induction

This technique takes a set of examples to try and generate general rules.

Advantages of rule induction

 If the knowledge is very complex and/or the domain is large then it may not be possible to develop expert
systems. If there is a 'bank' of old cases then rule induction may allow development of a system.
 The developer of a system does not have to have all the skills of a knowledge engineer.
 This approach may well allow for new knowledge to be acquired.

Knowledge Based Systems ~ Wainaina Page 3 of 9

  Once a base set of rules have been developed using rule induction these can be modified by the
knowledge engineer and expert together.

Disadvantages/Problems with rule induction

 Challenges on how are the attributes chosen
 Algorithms are inefficient.
 Only suitable for rule based systems
 Challenges on how to choose the training set
 The algorithms cannot usually deal with exceptions

Other Automatic Knowledge Acquisition Tools

i). Auto-Intelligence - A PC based tool that allows interactive KA with some rule induction
ii). Various Hybrid Approaches - Use of automatic tools and tools to aid the knowledge engineer.
iii). The Knacq - A methodology for KA 'by exception' that provides a tool for entering the knowledge and
automatic code generation for various shells.

Knowledge Acquisition Modes

Presently, the knowledge elicitation modes could be divided into three categories:
 Manual mode
 Automated mode
 Combined manual and automated mode

i) Manual Mode
This mode requires a direct interaction between the knowledge engineer and expert.

ii) Automated Mode

A number of software products exist in the market that is designed for automatic elicitation of
knowledge. The automated mode may be divided into two categories:
 Automated mode with the expert: - The expert interacts with the computer. The software is designed
to help the expert elicit knowledge without the assistance of a knowledge engineer.
 Automated mode without the expert: - The system is fed examples to produce rules or decision trees.

iii) Combined Manual and Automated Mode

 At the start of the process, the KE could use manual-based methods for conceptualization,
modularization, and gaining insights into the knowledge domain.
 In some of the modules for which enough examples are available and the accessibility to the expert is
limited, the knowledge engineer can use the automated mode to accelerate the elicitation of the
knowledge.
 Later, the expert should review the results, and may make suggestions to modify the automatically
developed rules.

Issues with Knowledge Acquisition

 Hard to get experts to express how they solve problems
 Representation on machine requires detailed expression i.e. at a very low level. Must be represented in a
structured way.
 Bringing together the ideas of all those involved in the knowledge transfer process. Many participants
are involved and they have varied backgrounds causing communication challenges (Expert, System
designers, Users, etc.).
 Mismatch between the way experts hold their knowledge and the way computers represent knowledge.

Various attempts have been made at overcoming these difficulties. This forms an important and active research
area. One approach is the use of natural language interfaces so that experts can communicate directly.
Computer aided knowledge acquisition tools are an alternative automated solution.

Knowledge Based Systems ~ Wainaina Page 4 of 9

VALIDATION AND VERIFICATION
These are terms used quality of a knowledge base. The methods are collectively known as evaluation
techniques.
i). Validation
This deals with the performance of the system i.e. how well it is making conclusions. The common
method of evaluation is the TURING TEST.
ii). Verification
This deals with the building of the system to the correct specification.

EXPERT SYSTEMS
Expert system is a set of computer programs that mimic the human expert. The programs will take and solve
problems in restricted problem domains or areas of expertise. Expert systems must therefore have
knowledge similar to the ones held by human experts and use it to solve problems of the type solved by
human beings.

STRUCTURE OF AN EXPERT SYSTEM

Knowledge base
The knowledge base contains the know-how of the human experts in a particular field. Such know-how is of
two types:-
 Facts: also known as "deep knowledge". These are based on generalized learning from School and
books and are well known and widely accepted by experts in the field.
 Heuristics: known as "surface knowledge". Heuristics consist of "rules of thumb", i.e. knowledge
attained through experience.
Knowledge base is derived and implemented from knowledge analysis and representation. It is the result of
knowledge acquisition and representation. For example, if you are using CLIPS to develop a rule-based
production system, the knowledge base will be presented in the form of production rules.

Inference engine is the brain of the expert system and, which has two primary tasks:
 Inference: the inference engine employs reasoning to examine existing facts and rules. Updates stored
knowledge and draws conclusions.

Knowledge Based Systems ~ Wainaina Page 5 of 9

  Control: the inference engine's mechanism for controlling the search of the knowledge base. A time
consuming task if the knowledge base is large and complex.

Working memory (Blackboard) is an area of the Computer's RAM reserved for storing information
regarding the current status of the problem. The blackboard may contain a plan of how the problem should
be solved, an agenda of potential actions awaiting execution and a solutions indicating candidate hypotheses
and alternate courses of action.

External data source: Some expert systems receive input data from external sources other than the user.
Common forms of this data - known as sensor data - include X rays, audible sounds and visual images. The
system interprets the data and makes inferences based on that data. Consider a medical system used to
monitor a patient's blood pressure. If the blood pressure reaches a certain critical level the system will
automatically take some action, perhaps sending an alarm signal to the user (the nurse).

User interfaces the part that enables the users or builders to submit their items to the expert system and also
the expert system to respond to the users or builders.

Explanation subsystem (justifier) is the part that traces responsibility for conclusions to their sources. It
may explain:
 Why some questions are asked by the expert system;
 How some conclusions are reached;
 Why some alternatives are rejected;
 The plan used to reach the solution;
 The remaining facts to establish before the final conclusion.

Knowledge refining system is the part that enables the analysis and use of knowledge so that learning may
take place and improvements may be made.

Participants in Expert Systems

The people involved in developing an expert system are: - The Knowledge Engineer(s), expert(s), user(s),
programmers and management.
The people are involved in various stages of the development process and interact heavily. It should be noted
here that one person can fulfil more than one role. For instance, an expert might also build the expert system
and, indeed, is a member of the Management team but this is not usually an ideal situation.

i). The Knowledge Engineer: The Knowledge Engineer is the person who actually develops expert
systems. This person requires skills and expertise that are peculiar to the world of expert systems.
Obviously a knowledge engineer needs to have a good technical grasp of expert systems software. This
entails understanding the intricacies of particular software as well as experience in programming in a
different 'style' to that of more conventional programming. This can obviously, to a large degree, be
taught but a high level of skill is only acquired by experience.
The job of the knowledge engineer involves very high interaction with people - the experts, users and
management. This means that he or she requires very good communication skills. The ability to talk at
the right level using tact and diplomacy is a very rare skill. It can be compared to a systems
analyst/programmer role with extra requirements of communication that extend beyond a systems
analyst.

ii). The Expert: The expert (or experts) is the person who provides most of the knowledge that is to be
encapsulated in the expert system. He or she will be the final arbiter on the correctness of the finished
system. The expert has experience or knowledge about a particular subject area (domain) and the
expert is chiefly responsible for providing domain knowledge.
The ideal expert is one who is: - articulate, enthusiastic and computer - literate
If an expert is not both articulate and enthusiastic there will be serious problems acquiring the
knowledge for the system. Computer literacy is ideal but not necessary as if he/she has an
understanding of what computers can and cannot do it will greatly enhance the discussions.

Knowledge Based Systems ~ Wainaina Page 6 of 9

iii). The User: The User (or, more usually, users) is the person who is actually going to use the finished
system. If the expert system is a consultative or advisory system he/she will interact heavily with the
finished system. If the system is not primarily consultative (e.g. plant control) the user will play less of
a role in development but will still need to be consulted at various stages. Problems occur when
individuals in a group of users have different skills and capabilities and care should be taken to ensure
that the target user is well defined.

iv). Management: Any expert system development that is to be successful has to have full support of
management at all levels. Expert systems technology is a relatively new one. A good deal of ignorance
and misinformation regarding their capabilities can lead to wariness amongst management to commit
scarce resources to an expert systems project. Therefore, management should be aware of expert
systems their potential benefits and realistic costs. The development team may, therefore, consider
giving a one day course or a day seminar to keep the management informed and maintain support and
resources.

Classic Problem Areas Addressed by Expert Systems

Expert systems have been used in several typical problem areas. Several application systems have been
demonstrated in these areas.

Category Problem area addressed

Interpretation Infer situation descriptions from observations. e.g. Surveillance, image analysis,
signal interpretation (e.g. PROSPECTOR)
Prediction Infer consequences of given situations. e.g. Weather forecasting, traffic
predictions, demographics (e.g. PLANT)
Diagnosis Infer malfunctions from observations e.g. Medical, mechanical, electronic,
software diagnosis (e.g. MYCIN)
Design Configure objects under constraints e.g. Circuit layouts, building design, plant
layout (e.g. R1/XCON)
Planning Develop plans to achieve goals e.g. Project management, routing,
communications, financial plans (e.g. MOLGEN)
Monitoring Compare observations and plans, flag exceptions e.g. Air traffic control, fiscal
management tasks (e.g. NAVEX)
Debugging Prescribe remedies to malfunctions e.g. Mechanical and software
Repair Execute plans to administer prescribed remedy. Generally they identify
weaknesses in knowledge and appropriate remedies
Instruction Diagnose, debug, and correct student performance Generally, they identify
weaknesses in knowledge and offer appropriate remedies (e.g. SOPHIE)
Control Interpret, predict, repair and monitor system behavior e.g. Life support, artificial
environment (e.g. VENTILATOR MANAGEMENT ASSISTANT)

Benefits of Expert Systems

i) Increased productivity and output since expert systems work faster than humans.
ii) Decreased decision-making time as expert systems can make decisions faster.
iii) Increased process and product quality as errors can be significantly reduced.
iv) Capture of scarce resource as ES can store the expertise held by humans who may be taking long
to train.
v) Flexibility as ES can sense changing needs and advice accordingly, such as a product out of
production.
vi) Easier equipment operation where ES is used to operate a complex equipment.
vii) Operation in harzadous environments where ES is used where humans are not safe such as in
nuclear power plants, toxic environments.
viii) Accessibility to knowledge and help desks where ES is used to supply information and scarce
knowledge or support help desks.
ix) Increased capability of other computerized system where ES integrate with other systems to make
applications work faster or produce higher quality results.

Knowledge Based Systems ~ Wainaina Page 7 of 9

 x) Ability to work with incomplete or uncertain information where ES may use existing knowledge to
solve a problem even though some other facts may still be missing or uncertain.
xi) Provide training where ES is used by novices because of its contents in the knowledge base.
xii) Enhanced problem solving and decision making where ES integrates analysis and judgement of top
experts.
xiii) Improved decision quality as ES is reliable and does not become bored, tired or hold attitudes.
xiv) Knowledge transfer to remote locations as ES can hold specialist knowledge that may be used in
remote locations such as eye disease diagnosis and treatment system demonstrated in Algeria and
Egypt by World Health Organization.

Limitations of Expert Systems

The following problems are associated with expert systems:
 Knowledge is not always readily available;
 Difficulty in extracting expertise from humans;
 Variations in problem assessment by different experts;
 Human experts cannot abstract when under pressure;
 Only work well in a narrow domain;
 Experts may not always validate their conclusions;
 Experts may not always use understood vocabulary;
 Knowledge engineers are few and expensive;
 End-users may not trust expert system;
 Subjectivity and biases in knowledge transfer.
Success Factors for Expert Systems
Implementation of expert systems may succeed because of the following factors:
 Management support and user involvement;
 There must be a high level of knowledge;
 There must be at least one cooperative expert;
 The problem must be qualitative (fuzzy);
 The problem must be sufficiently narrow in scope;
 There should be a good ES shell (store & manipulate knowledge naturally);
 The should be a friendly user interface for novice users;
 The problems must be difficult enough to warrant the use of ES;
 There should be competent knowledge-based system developers;
 The ES should positively impact on end-users.

BUILDING EXPERT SYSTEMS

Software for Building Expert Systems
Expert System tools: Software that is used for constructing expert systems. The tools range from programs
used for building expert systems to programs that can aid the knowledge acquisition process. The main
software tools for developing expert system fall into the following categories.
 Development software-programming languages (AI languages (Prolog and Lisp) and general purpose
languages); shells (e.g. JESS, KAPPA PC, CLIPS) and AI toolkits
 Development support tools- I/O facilities; debugging aids; explanation facilities; Editors
 Systems building tools-e.g. rule induction engines

Example
Write an expert system to diagnose the problem of a car that won’t start.

Solution
%Problem: Write an Expert system tha can help diagnise
%for a problem of a car that won’t start
%Diagnostic session: System probes for possible causes
%(Asks,notes response and give Recommendation)

car_wont_start:-
problem_with_battery;

Knowledge Based Systems ~ Wainaina Page 8 of 9

 problem_with_cables;
problem_with_starter;
problem_with_switch;
problem_with_fuel.
car_wont_start:-nl,nl,
write('I sorry I can''t diagnose the car problems ,maybe you never answered the
question correctly'),nl,nl.

problem_with_battery:-nl,
write('Is the battery voltage ok(Yes/No)'),
read(Battery),Battery=no,nl,nl,
write('Recharge battery and turn the KEY'),nl.

problem_with_cables:-nl,
write('Are the cables faulty(Yes/No)'),
read(Cables),Cables=yes,nl,nl,
write('Replace cables and turn the KEY'),nl.

problem_with_starter:-nl,
write('Is you starter motor ok(Yes/No)'),
read(Starter),Starter=no,nl,nl,
write('Do you want to repair or replace the motor?'),nl,nl,
write('=============================='),nl,nl,
write('1: Replace'),nl,
write('2: Repair'),nl,
write('Enter your Choice:'),
read(Choice),starter(Choice).

starter(Choice):-Choice=:=1,nl,write('Replace the motor with new one'),nl.

starter(Choice):-Choice=:=2,nl,write('Repair the starter and find out if it works'),nl.

problem_with_switch:-nl,
write('Is your ignition switch ok(Yes/No)'),
read(Switch),Switch=no,nl,nl,
write('Replace the switch and turn the KEY'),nl.

problem_with_fuel:-nl,
write('Is your petrol tank empty(Yes/No)'),
read(Fuel),Fuel=yes,nl,nl,
write('Buy petrol and turn the KEY'),nl.

EXPERT SYSTEM SHELLS

An expert system shell is a tool that has been specifically designed to enable speedy development of expert
systems. Building expert systems by using shells offers significant advantages. A system can be built to
perform a unique task by entering into a shell all the necessary knowledge about a task domain. The
inference engine that applies the knowledge to the task at hand is built into the shell. If the program is not
very complicated and if an expert has had some training in the use of a shell, the expert can enter the
knowledge himself. The following are key features of expert systems shells:
 The knowledge base and inference mechanism are totally separate.
 Shells offer a high level easy to use language for encoding the knowledge.
 The knowledge engineer does not have to write the inference engine since this is supplied with the tool.
 The development of a sophisticated user interface is usually facilitated by providing good screen designers.
 Most modern shells have tools such as editors, file handling facilities, debugging facilities and so on.

The User Interface

Since the user interface of expert systems is so important most shells offer facilities for developing an
effective user interface.
Facilities are usually available for developing good explanation and help facilities 'How' and 'Why' facilities
are usually built into the shell

Examples
There are literally hundreds of expert system shells available. Some examples are: CLIPS, JESS, Mycin,
Babylon, G2, etc

Knowledge Based Systems ~ Wainaina Page 9 of 9