1

EXPERT SYSTEMS NOTES

(A). D E F I N I T I O N
Systems that use human knowledge captured in computers to solve problems that
ordinarily require human expertise. Human experts include medical doctors, computer
hardware engineers, vehicle mechanics, economists, weather forecasters, geologists, etc.

(B). C O M P O N E N T S
(i). Knowledge Acquisition System:
System for accumulation, transfer, and transformation of problem solving expertise
from experts or documented sources into a computer program.

Sources of Knowledge: human experts, text books, multimedia documents,

databases, research reports, web, etc.

(ii). Knowledgebase
Contains the knowledge necessary for understanding, formulation, and solving
problems.
Elements: Facts (such as problem situation and theory of the problem area), and
rules (that direct the use of the knowledge to solve problems).

(iii). Inference Engine

The ‘brain’ of the expert system. It’s the control structure or rule interpreter. It’s
a program that provides a methodology for reasoning about the information in the
knowledgebase.

Components
(I). The interpreter: The rule interpreter.
(II). A scheduler: Maintains control over the agenda.
(III). A consistency enforcer: Attempts to maintain a consistent representation of
the emerging solution.

(iv). User Interface

A language processor. It provides a friendly, problem oriented communication
between the user and the computer.

(v). Blackboard (Workplace)

An area in the working memory set aside for the description of the current problem.

(vi). Explanation (Justifier)

Explain why conclusions have been reached, why a question has been asked, why
some alternatives were rejected, etc.

(vii). Knowledge Refining System

It analyzes the existing knowledge, learns from it, and improves it for future
consultation.
 2

(C). P E R S O N S C O N C E R N E D I N E X P E R T S Y S T E M S
(i). Expert
The human being with extensive knowledge, judgment, experience and methods
along with the ability to apply these talents to give advice and solve problems.

(ii). Knowledge Engineer

Helps the expert structure the problem oriented area by interpreting and intergrating
human answers to question, drawing analogies, posing counter examples, and
bringing to light conceptual difficulties.
Also the system builder.

(iii). The User

The non-expert seeking advice, a student who wants to learn (The ES is the
instructor), An ES builder who wants to increase the knowledgebase.

(iv). Other Participants

System builder (to integrate the ES with other systems), tool builder (Builds
specific tools), Vendors (Provides tools and advice), support staff (Provides clerical
and technical help).

(D). P R O B L E M A R E A S
(i). Interpretation Systems: Explains observed data.
E.g. Surveillance, speech understanding, image analysis, signal interpretation.

(ii). Prediction Systems: Infer likely consequences of given situations.

E.g. Weather forecasting, demographics predictions, economic forecasting, traffic
predictions, crop estimates, military, marketing, financial forecasting.

(iii). Diagnostic Systems: Infer system malfunctions from observations.

E.g. Medical, electronic, mechanical, software diagnostic systems.

(iv). Design systems: Configure objects under constraints.

(v). Planning Systems: E.g. project planning, management, routing, communications,

product development, military application, financial planning.

(vi). Monitoring Systems: E.g. Air control, fiscal management.

Etc.

(E). C O M P A R I S O N W I T H C O N V E N T I O N A L P R O G R A M S

Conventional Programs Expert Systems (AI programs e.g.

Prolog)
Information, processing usually combined Knowledgebase clearly separate from the
 3

in one sequential program. processing (inference)

Program does not make mistakes Program may make mistakes
Do not (usually) explain why input data is Explanation is part of most expert systems
needed or why conclusions reached
Require all input data. May not function Do not require all initial facts. Typically
properly with missing data can arrive at reasonable conclusion with
missing facts
Changes in program are tedious Changes in rules are easy to make
Program operates only when complete Program can operate with only a few rules
Execution done step by step Execution done by using rules (heuristics)
and logic
Effective manipulation of large databases Effective manipulation of large knowledge
bases
Representation and use of data Representation and use of knowledge
Efficiency is the major goal (e.g. program Effectiveness is the major goal (e.g. how
should be as small as possible) good an advice/conclusion is)
Easily deal with quantitative data Easily deal with qualitative data
(relationships) e.g. identifying patterns
Use numerical data representations Use symbolic knowledge representations

(F). BENEFITS of Expert Systems

 Availability − They are easily available due to mass production of software.
 Less Production Cost − Production cost is reasonable. This makes them affordable.
 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.

(G). Development of Expert Systems: General Steps

The process of ES development is iterative. Steps in developing the ES include −
(i) Identify Problem Domain
The problem must be suitable for an expert system to solve it.
Find the experts in task domain for the ES project.
Establish cost-effectiveness of the system.

(ii) Design the System

Identify the ES Technology
Know and establish the degree of integration with the other systems and databases.
Realize how the concepts can represent the domain knowledge best.

(iii) Develop the Prototype

 4

From Knowledge Base: The knowledge engineer works to −

Acquire domain knowledge from the expert.
Represent it in the form of If-THEN-ELSE rules.

(iv) Test and Refine the Prototype

The knowledge engineer uses sample cases to test the prototype for any deficiencies in
performance.
End users test the prototypes of the ES.

(v) Develop and Complete the ES

Test and ensure the interaction of the ES with all elements of its environment, including
end users, databases, and other information systems.
Document the ES project well.
Train the user to use ES.

(vi) Maintain the System

Keep the knowledge base up-to-date by regular review and update.
Cater for new interfaces with other information systems, as those systems evolve.

(F). E X E R C I S E S
(i). Explain the meaning of the term expert system. Describe its components.

(ii). Describe four types of problems that are solved by expert systems.

(iii). Explain the human elements in expert systems.

(iv). Explain five differences between expert systems and conventional programs.
 5

NATURAL LANGUAGE PROCESSING

(NLP) NOTES

(A). D E F I N I T I O N
Research aimed at making people and computers communicate in a natural (human) language such as
English rather than in computer language.

Two fields:
(i). Natural language understanding: Making computers understand English statements so that computers
can communicate with people more easily.

(ii). Natural language generation: Making computers produce English statements.

(B). A P P L I C A T I O N S
 Machine translation E.g. newspaper translation. E.g. SPANAM (a system that translates between
English and Spanish), Taum-Meteo (developed at university of Monstreal to translate weather
reports from English to French).
 Database access: Allows user to interact with the database without having to learn a formal language
like SQL.e.g. Fernando Pereira’s CHAT developed in 1983. it responds to queries directed to a
geographical database. It answers questions such as which countries are bordered by two seas? What
are the countries from which a river flows into the black sea?
 Information retrieval: Retrieving documents that are relevant to a query from a group of documents.
The query is a list of words describing the contents of the required document.
 Data extraction: Extracting useful data from a natural language text. E.g. SCISOR system
(developed 1990) extracts various pieces of data from a news story.

(C). C H A L L E N G E S
(i). Ambiguity: Many things we say can be interpreted in many ways. Includes;
(I). Multiple word meanings. E.g. the word “Pen”.
(II). Syntactic ambiguity: E.g. “I hit the man with a hammer”.
Question: Picked a hammer and hit the man or hit the man who was holding a hammer?
(III). Categorical ambiguity: E.g. The word “Time” can be a noun, a verb or an adjective. E.gs.
Sentences: “Time is money” (Noun), “Time me on the last lap” (Verb), “Time travel is not
likely in my lifetime” (adjective).
(IV). Referential ambiguity: More than one objects can be referred to by a noun phrase.
E.g. “After they finished the exam, the students and the lecturers left”. Who finished
the exam (students or the lecturers)?
(V). Unclear ante cents: (use of pronouns). E.g. John hit Bill because of his failure to
understand”. Who failed to understand, John or Bill?
(ii). Incompleteness: We do not always say all what we mean. Because we share
common experiences, we usually leave out many details and assume that our
listeners can ‘read between the lines’.
(iii). Inaccuracy: Includes spelling errors, ungrammatical constructions, and
improper punctuation.

(iv). Imprecision: People often express concepts in vague and in inexact

terminology: E.g. how long is “long time”?
 6

Resolving the Challenges: Techniques

(i). Put the idea in Context i.e. relate ambiguous sentence to what the paragraph is
talking about.
(ii).Use familiarity principle to relate the idea with familiar situations. E.g. “The student
took too long to answer the exam.
(iii). Infer the intended meaning by recognizing familiar patterns to solve the problem
of inaccuracy.
 7

(D). N L P U N D E R S T A N D I N G T E C H N I Q U E S
Steps taken to try and understand the meaning of sentences.

(i). lexical analysis

Words are looked up in a dictionary to determine their meanings.

(ii). Syntactic Analysis

Conducting an analysis of a sentence’s syntax (form).
Uses a parsing technique (a method of separating the sentence into its component
parts so as to analyze them).
It shows the role of each word in the sentence.
E.g. “the man writes” will pass this technique but “the men writes” will not.

A sentence is made up of a noun part followed by a verb part. A singular noun is

followed by a singular verb. So are plurals.

(iii). Semantic analysis

Deals with understanding the meaning of a sentence rather than its form (as in
syntactic).
E.g. If the sentence “the man eats the orange” passes syntactic analysis, then the
sentence “the orange eats the man” will also pass the analysis but will fail in
semantic analysis.

(iv). Pragmatic analysis

Deals with how the language is used. It shows how people really mean in their
sentences.
E.g. the sentence “can you open the door” could be a question (for above three
techniques) or a polite request (for pragmatic technique).

(E). W A Y S O F R E P R E S E N T I N G NLP SYNTAX

(i). Use of parse trees: Old method

(ii). Use of transition networks: Old method

(iii). Use of lists.

Current method
E. g. to represent “sue hit tom”

[s, [np, [vp]],

[noun, tom]
[noun, sue]
[verb, hit]

[s, [np, [proper_noun, sue]], [vp, [v, hit], [np, [proper_noun, tom]]]]
 8

(F). E X E R C I S E S
(i). Discuss areas where you think NLP could be applied in Kenya.
(ii). Describe with examples the reasons to why NLP is a big problem to researchers.
(iii).Explain the ambiguities in the following sentence.
(I). Patrick argued with Alex in his office.
(II). Here you are.
(III). Go to train.

What approach(es) that can be used to solve the above ambiguities?