ADVANCED

KNOWLEDGE ENGINEERING
Instructor:
KHUAT Thanh Tung
University of Technology Sydney
Subject’s Outline

Ø Topic 1: An Overview of Knowledge Engineering

Ø Topic 2: An Overview of Knowledge-based Systems
Ø Topic 3: Knowledge Acquisition
Ø Topic 4: Knowledge Representation and Reasoning
Mid-term assessment
Ø Topic 5: Ontology
Ø Topic 6: Knowledge Graphs
Ø Topic 7: Expert Systems
Ø Topic 8: Uncertain Reasoning
Ø Topic 9: Hybrid Knowledge-based Systems
Ø Topic 10: Automated AI Planning
Group projects for the advanced topics 2
Subject’s objectives
Ø Understanding the Knowledge Base, Knowledge-based
Reasoning and Representation
Ø Learn how to build fundamental components of
Knowledge-based systems
Ø Learn how to integrate the knowledge base, and
knowledge-based reasoning components into Machine
Learning Systems
Ø Figure out several advanced topics related to Knowledge
Engineering: Ontology, Knowledge Graphs, Hybrid
knowledge-based systems, and Automated AI Planning
using knowledge base
Ø Programming languages: General purpose programming
language: Python, C++, Java,… + PROLOG, PDDL
3
Course outcomes
Ø Remembering: Techniques for knowledge acquisition,
representation, reasoning, and KBS development.
Ø Understanding: Fundamental components of Knowledge-
based systems, methods of building a specific KBS
Ø Applying:
Ø Have the ability to build small applications using knowledge and
related components
Ø Able to use several programming languages to build core
reasoning engines within a knowledge-based system such as
PROLOG
Ø Analyzing:
Ø Analyze the given problems and choose suitable techniques,
knowledge presentations and a KBS
Ø Able to integrate the knowledge engineering into machine
learning and AI systems. 4
Assessment methods

Ø Learning progress assessment (20%):

Ø Random attendance checking
Ø Small exercises
Ø Mid-term: A set of quizzes based on the lectures (20%)
Ø Final assignment (60%): A group project of 3-5 students
for provided topics.

5
 References

1. Textbook: An Introduction to Knowledge Engineering, written

by S.L. Kendal and M. Creen, published by Springer
2. An Introduction to Knowledge Graphs, written by Umutcan
Serles and Dieter Fensel, published by Springer
3. Logic Programming with Prolog, written by Max Bramer,
published by Springer
4. Knowledge Engineering Tools and Techniques for AI Planning,
edited by Mauro Vallati and Diane Kitchin, published by Springer
5. Ontologies - A Handbook of Principles, Concepts and
Applications in Information Systems, edited by Raj
Sharman, Rajiv Kishore, Ram Ramesh
6. Giáo trình các hệ cơ sở tri thức, biên soạn bởi GS. Hoàng Văn
Kiếm, TS. Đỗ Phúc, TS. Đỗ Văn Nhơn, xuất bản bởi Nhà xuất bản
đại học quốc gia thành phố Hồ Chí Minh
6
 AN OVERVIEW INTRODUCTION
TO KNOWLEDGE ENGINEERING

TOPIC 1
Objectives of this topic

By the end of this topic, you will be able to:

ü define knowledge and explain its relationship to data and
information
ü distinguish between knowledge management and
knowledge engineering
ü explain the skills required of a knowledge engineer
ü define Knowledge-based systems
ü explain what a KBS can do

8
Agenda
¨ What is knowledge?
¨ Types of knowledge
¨ Knowledge engineering
¨ Knowledge engineers

9
Philosophical Basis
¨ Traditional questions that have been analyzed by
philosophers, psychologists, and linguist:
¤ What is knowledge?
¤ What do people have inside their head when they know
something?
¤ Is knowledge expressed in words?
¤ If so, how could one know things that are easier to do than to say,
like tying a shoestring or hitting a baseball?
¤ If knowledge is not expressed in words, how can it be transmitted
in language?
¤ How is knowledge related to the world?
¤ What are the relationships between the external world,
knowledge in the head, and the language used to express
knowledge about the world?

10
Philosophical Basis
¨ With the advent of computers, the questions
addressed by the field of artificial intelligence (AI)
are:
¤ Can knowledge be programmed in a digital computer?
¤ Can computers encode and decode that knowledge in
ordinary language?
¤ Can they use it to interact with people and with other
computer systems in a more flexible or helpful way?

11
Information Processing Views of
Knowledge
¨ Hierarchical view: data information knowledge
¤ Data are streams of raw facts representing events before they
have been arranged into a form that people can understand and use
¤ Information is the input or raw material of new knowledge
¤ Knowledge is authenticated/personalized information
¨ Reversed hierarchical view: knowledge information data
¤ Knowledge must exist before information can be formulated and
before data can be collected
¨ Non-hierarchical view: data information

Knowledge
¤ Knowledge is needed in converting data into information
¤ Knowledge is the accumulation of experiences vs. knowledge is
created through conjectures and refutations.

12
Information Processing Views of
Knowledge
¨ The movement from data to knowledge implies a shift from facts
and figures to more abstract concepts
¨ From a knowledge engineering perspective, knowledge can be
considered as something that can be expressed as a rule or useful to
assist a decision

13
Alternative Perspectives on Knowledge
¨ Knowledge can be defined as a justified belief that
increases an entity’s capacity for effective action.
¨ It may be viewed from several perspectives:
(1) a state of mind – knowledge is the state of knowing and
understanding
(2) an object – knowledge is an object to be stored and
manipulated
(3) a process – knowledge is a process of applying expertise
(4) a condition – knowledge is organized access to and
retrieval of content
(5) a capability – knowledge is the potential to influence
action

14
Taxonomies of Knowledge
¨ Tacit vs. explicit
¤ Explicit knowledge refers to knowledge that is
transmittable in formal, systematic language
¤ Tacit knowledge is deeply rooted in actions, experience,
and involvement in a specific context. It consists of cognitive
element (mental models) and technical element (know-how
and skills applicable to specific work).
¨ Individual vs. social
¤ Individual knowledge is created by and exists in the
individual whereas social knowledge is created by and
exists in the collective actions of a group.

15
Taxonomies of Knowledge
¨ Five Types of Knowledge
¤ Declarative knowledge
n Know-about
¤ Procedural knowledge
n Know-how
¤ Causal knowledge
n Know-why
¤ Conditional knowledge
n Know-when
¤ Relational knowledge
n Know-with
¨ Meta-knowledge
¤ Knowledge about knowledge

16
 Four Modes of Knowledge Conversion
(Nonaka, 1991)
Tacit knowledge Explicit knowledge
To

Tacit
Socialization Externalization
knowledge

From

Explicit Internalization Combination

knowledge

Socialization is the process of sharing tacit knowledge through observation,

imitation, practice, and participation in formal and informal communities.
Externalization is the process of articulating tacit knowledge into explicit concepts
Combination is the process of integrating concepts into a knowledge system.
Internalization is the process of embodying explicit knowledge into tacit knowledge 17
Knowledge Engineering

¨ An engineering discipline that involves integrating

knowledge into computer systems in order to solve
complex problems normally requiring a high level of
human expertise (Feigenbaum and Pamela, 1983)

¨ It normally involves five distinct steps in transferring

human knowledge into some form of knowledge
based systems (KBS)

18
Five Steps of Knowledge Engineering

¨ Knowledge acquisition
¨ Knowledge validation

¨ Knowledge representation

¨ Inferencing

¨ Explanation and justification

19
Five Steps of Knowledge Engineering

20
Two Main Views of Knowledge
Engineering
¨ Transfer view – This is the traditional view. In this
view, the key idea is to apply conventional
knowledge engineering techniques to transfer
human knowledge into the computerized system.
¨ Modeling view – In this view, the knowledge
engineer attempts to model the knowledge and
problem solving techniques of the domain expert
into the computerized system.

21
Knowledge Engineering (KE) vs.
Knowledge Management (KM)
¨ KE is primarily concerned with constructing a
knowledge-bases system while KM is primarily
concerned with identifying and leveraging
knowledge to the organization’s benefit.
¨ KE and KM activities are inherently interrelated.

¨ Knowledge engineers are interested in what

technologies are needed to meet the enterprise’s
KM needs.

22
Knowledge Engineers
¨ A knowledge engineer is responsible for obtaining knowledge
from human experts and then entering this knowledge into
some form of KBS.
¨ In developing KBS, the knowledge engineer must apply
methods, use tools, apply quality control and standards, plan
and manage projects, and take into account human, financial,
and environmental constraints.
¨ Required skills of a knowledge engineer
¤ Knowledge representation
¤ Fact finding (knowledge elicitation)
¤ Human skills
¤ Visualization skills
¤ Analysis
¤ Creativity
¤ Managerial
23
 An example of a KBS Project
qThe goal: assist the clinician in the intensive care unit (ICU)
qProblems need to be addressed:
q interpretation of measurement values with respect to historical
information about changes in a patient’s status and therapy.
q The difficulty of directly relating measurement values to a
therapeutic recommendation
q Tasks of the system:
q Predict the initial setting of the mechanical ventilator to assist
the patient to breathe
q Suggest adjustments to treatment by continuous reassessment
of the patient’s condition
q Summarize the patient’s physiological status
q Maintain a set of patient’s specific expectations and goals for
future evaluations
q Aid in the stabilization of the patient’s condition 24
 An example of a KBS Project
qKBS development process

25
 An example of a KBS Project
qOne of the main queries in the project was from the experts
providing knowledge for the system
q it was essential that the system provided accurate answers,
otherwise patients lives could be at risk
qExperts providing the knowledge did not want to be blamed if
an incorrect response was given by the KBS

26
 An example of a KBS Project

qThe knowledge engineers and project managers need to provide

the main assurance that the built system is in accordance with
quality assurance standards
qQuality assurance is an essential part of the design of any KBS,
especially
q railway signaling systems
q alarm systems
q detection of gas leaks
q nuclear power station monitoring and control
q Other critical systems in which an error could result in significant
risk, including loss of life

27
 An example of a KBS Project

qThe Project Manager’s Dilemma

qLike the general software project management, a project
manager for the KBS project needs some negotiating skills to
try and match the expectations of all parties involved in a
project

28
 KNOWLEDGE-BASED SYSTEMS

TOPIC 2
29
Objectives of this topic

By the end of this topic, you will be able to:

ü describe the characteristics of a knowledge-based system
ü explain the main elements of knowledge-based systems
and how they work
ü evaluate the advantages and limitations of knowledge-
based systems
ü identify appropriate contexts for the use of particular
types of knowledge-based systems
ü distinguish expert systems from machine learning systems
ü state a brief definition of expert systems, neural networks,
case-based reasoning, genetic algorithms, intelligent
agents
30
Agenda
¨ Expert systems
¨ Neural networks

¨ Case-based reasoning

¨ Genetic algorithms

¨ Intelligent agents

31
What are KBSs?
¨ A knowledge based system is a system that uses
artificial intelligence techniques in problem-solving
processes to support human decision-making, learning,
and action.
¨ Two central components of KBSs are
¤ Knowledge base
n Consists of a set of facts and a set of rules, frames, or procedures
¤ Inference engine
n Responsible for the application of knowledge base to the problem on hand.

¨ There are pros and cons of using KBSs, compared to

human expertise.
32
Types of KBSs
¨ Expert systems
¨ Neural networks

¨ Case-based reasoning

¨ Genetic algorithms

¨ Intelligent agents

33
Expert Systems
¨ An expert system is a computer program designed
to emulate the problem-solving behavior of an
expert in a specific domain of knowledge
¨ In order to qualify as an expert system, a system
must have the capability of explaining or justifying
its conclusions.
¨ A system which can explain its reasoning process is
said to demonstrate meta-knowledge (knowledge
about its own knowledge).

34
Features of Problem Solvers
¨ Human experts exhibit certain characteristics and
techniques which help them perform at a high level
in solving problems in their domain.
q Solve the problem
q Explain the results
q Learn
q Restructure knowledge
q Break rules
q Determine relevance
q Degrade gracefully

35
Characteristics of Expert Systems
q The system performs at a level generally recognized as
equivalent to that of a human expert or specialist in the
field.
qThe system is highly domain specific.
qThe system can explain its reasoning.
qIf the information with which it is working is
probabilistic or fuzzy, the system can correctly
propagate uncertainties and provide a range of
alternative solutions with associated likelihoods.

36
Applications of Expert Systems
¨ DENDRAL
¤ Applied knowledge (i.e., rule-based reasoning)
¤ Deduced likely molecular structure of compounds
¨ MYCIN
¤ A rule-based expert system
¤ Used for diagnosing and treating bacterial infections
¨ XCON
¤ A rule-based expert system
¤ Used to determine the optimal information systems
configuration
¨ New applications: Credit analysis, Marketing, Finance,
Manufacturing, Human resources, Science and
Engineering, Education, …
37
Components of Expert Systems
¨ Knowledge base
¤ Consists of facts and rules
¤ Rules are commonly expressed in if-then structure (production rules)
n If-premise then conclusion
n If-condition then action
¨ Inference engine
¤ Responsible for rule interpretation and scheduling
¤ Forward chaining vs. backward chaining
¨ User interface
¨ Working memory
¨ Explanation facility

38
Conceptual Architecture of a
Typical Expert System
Modeling of Manufacturing Systems

Abstract

ajshjaskahskaskjhakjshakhska akjsja s
askjaskjakskjas

Expert(s)
Printed Materials

Expertise Information

Knowledge
Control Engineer Structured
Structure Knowledge

Inference
External Engine Knowledge Knowledge
Interfaces Base(s)
Working
Memory

Base Model Questions/

Answers
Data Bases
Spreadsheets Solutions Updates

User
Interface
39
Expert System Building Tools
¨ Programming language
¤ An expert system can be implemented using a general
purpose programming language. However, the
programming language LISP and PROLOG are typically
used in expert systems implementation, in particular
Artificial intelligence applications.
¨ Shells
¤ A shell consists mainly of an inference engine and an editor
to assist developers in building their knowledge base.
¤ Example: CLIPS is an expert system shell developed by
NASA
40
Strengths and Limitations of Expert
Systems
qStrengths
q Human expertise can be expensive
q Expert Systems only require the costs of hardware to run
q Human advice can be inconsistent
q Human advice may be adversely affected by tiredness, busy
diaries, personal problems, etc.
q Computer advice will always be based on the rules within the
expert system, and those rules can be checked by other experts to
ensure their validity.
q Human knowledge may be lost
q humans tend to die eventually
q Human knowledge may be lost in other ways such as brain disease
or simply changing jobs
q Human knowledge can only be accessed in one place at one time
q where the expert happens to be
41
Strengths and Limitations of Expert
Systems
qLimitations
q Lack of common sense
q humans may draw conclusions based on their overall view of the
world; expert systems do not have this information
q Lack of inspiration or intuition
q Lack of flexibility to apply their knowledge outside a relevant
domain
q Humans understand the limits of their knowledge and will seek help
when confronted by complex or novel situations
q Unless programmed specifically, expert systems will not recognise
their limitations and fail when confronted with new situations

42
Where Are Expert Systems Used?
qThe problem is important to business
q time or money or both can be saved by using the expert system
qThe expertise required is available and stable
q Able to build the expert system rule base
qThe knowledge required is scarce
q at least in terms of human experts available to provide answers
qThe problem is recurrent
qThe problem is at the right level of difficulty
qThe domain is well defined and of a manageable size
q Particularly large domains or domains with no easily defined
limits (e.g., object identification systems in the autonomous cars)
are difficult to program due to the large number of required rules
q The solution depends on logical reasoning, not ‘common sense’ or
general knowledge
43
 Machine Learning Methods
¨ Machine learning
The process by which a computer learns from experience
(e.g., using programs that can learn from historical cases)
Machine learning

Supervised Unsupervised Reinforcement

Neural Auto Dimensionality

Clustering
networks encoders reduction

Explanation Case-
Statistical Inductive Neural Genetic
based based
regression learning networks algorithms
learning reasoning 44
 An examples of Machine Learning

A self-driving car system uses dozens of components that

include detection of cars, pedestrians, and other objects
45
 An examples of Machine Learning

One way to build a detection system is to write down rules.

In practice, it's almost impossible for a human to specify all the

46
edge cases.
 An examples of Machine Learning
The machine learning approach is to teach a computer how to do
detection by showing it many examples of different objects.

No manual programming is needed: the computer learns what

defines a pedestrian or a car on its own! 47
A formal definition of Machine
Learning

Machine learning is a field of study that gives

computers the ability to learn without being
explicitly programmed.
(Arthur Samuel, 1959)

This principle can be applied to countless domains: medical

diagnosis, factory automation, machine translation, and many
more!
48
 Neural Networks

q Neural networks represent a brain metaphor for

information processing. Neural computing refers to a
pattern recognition methodology for machine learning. The
resulting model from neural computing is often called an
artificial neural network (ANN) or neural network (NN).

q Due to their ability to learn from the data, their non-

parametric nature (i.e., no rigid assumptions), and their
ability to generalize, neural networks have been shown to
be promising in many forecasting and business classification
applications.

49
Basic Concepts of Neural Networks
¨ The human brain is composed of special cells called nuerons.
¨ Neural network elements
¤ Nucleus
The central processing portion of a neuron
¤ Soma
The main body of the neuron in which the cell nucleus is contained
¤ Dendrite (sợi nhánh)
The part of a biological neuron that provides inputs to the cell
¤ Axon (sợi trục)
An outgoing connection (i.e., terminal) from a biological neuron
¤ Synapse
The connection (where the weights are) between processing elements
in a neural network 50
Structure of a Biological Neural
Network

51
 Artificial Neural Network

q An ANN model emulates a biological neural network

q Neural concepts are usually implemented as software
simulations of the massive parallel processes that
involve processing elements (also called artificial
neurons) interconnected in a network structure.
q Connections between neurons have an associated
weight.
q Each neuron calculates a weighted sum of the
incoming neuron values, transforms this input, and
passes on its neural value as the input to subsequent
neurons or external outputs.
52
Processing Information in an
Artificial Neuron

53
The Relationship Between Biological
and Artificial Neural Networks

Biological Ar tificial
Soma Node
Dendrites Input
Axon Output
Synapse Weight
Slow speed Fast speed
Many neurons (109) Few neurons (a dozen to
hundreds of thousands)

54
Neural Network with One Hidden
Layer

55
Example of ANN Functions

56
Learning in ANN
q Supervised learning
q Uses a set of inputs for which the desired outputs are known
q Example: Backpropagation algorithm

q Unsupervised learning
q Uses a set of inputs for which NO desired output are known
q The system is self-organizing; that is, it organizes itself
internally. A human must examine the final categories to assign
meaning and determine the usefulness of the results.
q Example: Self-organizing map

57
Learning in ANN
q Supervised learning
q Uses a set of inputs for which the desired outputs are known
q Example: Backpropagation algorithm

q Unsupervised learning
q Uses a set of inputs for which NO desired outputs are
known.
q The system is self-organizing; that is, it organizes itself
internally. A human must examine the final categories to
assign meaning and determine the usefulness of the results
q Example: Self-organizing map.

58
Characteristics of ANNs

¨ Adaptive learning
¨ Self-organization

¨ Error tolerance

¨ Real-time operation

¨ Parallel information processing

59
Benefits and Limitations of Neural
Networks
¨ Benefits
¤ Abilityto tackle new kinds of problems
¤ Robustness

¨ Limitations
¤Performs less well at tasks humans tend to find difficult
¤ Lack of explanation facilities

¤ Requires large amounts of training and test data

60
 Case-Based Reasoning (CBR)
q A case has two parts: a problem and a solution
q Cases represent experience; that is, they record how a
problem was solved in the past
q CBR is a methodology in which knowledge and/or
inferences are derived from historical cases. It is based on
the premise that new problems are often similar to
previously encountered problems and that, past solutions
may be of use in the current situations.
q CBR is particularly applicable to problems in which the
domain is NOT understood well enough for a robust
statistical model or system of equations to be formulated.
61
Process of CBR
1. Retrieve
§ Given a target problem, retrieve the most similar cases

2. Reuse
§ Map the solution and reuse the best old solution to solve the
current case

3. Revise
§ Test the solution and, if necessary, revise the old case to come
up with the solution

4. Retain
§ After the solution has been successfully adapted to the
target problem, store the resulting experience as a new
case 62
Step-by-Step Process of CBR

63
Similarity Computation

q Cases are ranked according to their similarity based on

the similarity of each feature
q The degree of similarity can be expressed by a real
number between 0 (not similar) and 1 (identical).
q The importance of different features may be different.
In that case, similarity is computed by weighted average.

64
CBR Examples
¨ Intelligent customer support and sales support
¨ Retrieval of tour packages from travel catalogs
¨ Conflict resolution in air traffic control
¨ Conceptual building design aid
¨ Conceptual design aid for electronic devices
¨ Medical diagnosis
¨ Aircraft troubleshooting
¨ Heuristic retrieval of legal knowledge
¨ Computer supported conflict resolution through negotiation or
mediation

65
Advantages and Disadvantages of
Using CBR
q Advantages
q Improved knowledge acquisition
q Reduced development time
q Easier explanation
q Learning over time

q Disadvantages
q Storing of cases in the Knowledge base
q Implicit link between problem and solution
q Access and retrieval speed.

66
Genetic Algorithms

q Programs that attempt to find optimal solutions to

problems by conceptually following steps inspired by
the biological processes of evolution
q The method learns by producing offspring that are
better and better, as measured by a fitness-to-survive
function, until an optimal or near-optimal solution is
obtained.

67
Genetic Algorithm Fundamentals

¨ Chromosome
A candidate solution for a genetic algorithm
¨ Fitness function
A measure of the objective to be obtained.
¨ Generation
An iteration of the genetic algorithmic process
in which candidate solutions are combined to
produce offspring
68
Processes within Genetic Algorithm

¨ Reproduction
¤ Through reproduction, genetic algorithms produce new generations
of improved solutions by selecting parents with higher fitness
ratings or by giving such parents a greater probability of being
contributors and by using random selection.
¨ Crossover
¤ The combining of parts of two superior solutions by a genetic
algorithm in an attempt to produce an even better solution
¨ Mutation
¤ A genetic operator that causes a random change in a potential
solution

69
Genetic Algorithm Process

70
Genetic Algorithm Parameters

qSome parameters must be for the genetic algorithm

q Number of initial solutions to generate
qNumber of offspring to generate
qNumber of parents and offspring to keep for the next
generation
qMutation probability
qProbability distribution of crossover point occurrence
q Their values are dependent on the problem being
solved and are usually determined through trial and
error

71
Genetic Algorithm Benefits and
Limitations
q Genetic algorithms are particularly useful for complex
problems that require rapid development of set of good
solutions

q Limitations
q Not all problems can be framed in the mathematical
manner that genetic algorithms demand.
q Development of a genetic algorithm is complex.
q In some situations, the “genes” from a few comparatively
highly fit (but not optimal) individuals may come to
dominate the population, causing it to converge on a local
maximum
q Most genetic algorithms rely on random number generators
that produce different results each time the model runs
72
 Genetic Algorithm Applications
q Genetic algorithms provide a set of efficient, domain-
independent search heuristics for a broad spectrum of
applications including
q Dynamic process control
q Complex design of engineering structures
q Scheduling
q Transportation and routing
q Layout and circuit design
q Telecommunications
q Discovery of new connectivity typologies

73
 Intelligent Agents

q A computer program that carries out a set of operations on

behalf of a user or another program, with some degree of
autonomy, and in doing so, employs some knowledge or
representation of the user’s goals or desires.
q Agents in various forms
q Software agents, wizards, software daemons
q e-mail agents (mailbots), web browsing assisting agents
q intelligent search agents (Web robots, spiders)
q Internet softbots, network management and monitoring agents
q e-commerce agents

74
 Features of Intelligent Agents
q Reactivity
q Agents perceive their environment and respond in a timely
fashion to changes that occur in it.
q Proactiveness
q Agents are able to exhibit goal-directed behavior by
taking initiative
q Social ability
q Agents are capable of interacting with other agents in
order to satisfy their design objectives.
q Autonomy
q Agents must have control over their own actions and be
able to work and launch actions independently of the user
or other actors 75
 Why Use Intelligent Agents
q The Gartner Group findings on information overload:
q The amount of data collected by large enterprises doubles
every year
q Knowledge workers can analyze only about 5% of this data
q Most of the knowledge workers’ efforts are spent in trying to
discover important patterns in the data (60% or more), a much
smaller percentage in determining what these patters mean
(20% or less), and very little time (10% or less) is spend
actually doing something about the patterns.
q Information overload reduces our decision-making capabilities
by 50 percent.
q A major value of intelligent agents is that they are able to
assist in searching through all the data.
q Intelligent agents save time by making decisions about what is
relevant to the user as well as by automating routine tasks. 76
Intelligent Agents: How Smart Are
They?
¨ Intelligence levels
¤ Level 0 - Agents retrieve documents for a user under
straight orders
¤ Level 1 - Agents provide a user-initiated searching
facility for finding relevant Web pages
¤ Level 2 - Agents maintain users’ profiles

¤ Level 3 - Agents have a learning and deductive

component to help a user who cannot formalize a query
or specify a target for a search

77
Intelligent Agents Vs. Expert Systems

¨ Agents and expert systems are similar in that they

both intend to incorporate domain knowledge to
automate decision making.
¨ They are different in the following aspects:
¤ Classic ESs are not coupled to any environment in which they
act; they act through a user as a middle man. Agents can
actively search information from the environment in which
they reside.
¤ ESs are not generally capable of reactive and proactive
behavior.
¤ ESs are not generally equipped with social ability in the
sense of cooperation, coordination, and negotiation.
78
Internet-based Software Agents
¨ Nine major application areas:
¤ Assisting in workflow and administrative management
¤ Collaborating with other agents and people

¤ Supporting e-commerce

¤ Supporting desktop applications

¤ Assisting in information access and management, including

searching and FAQs
¤ Processing e-mail and messages

¤ Controlling and managing network access

¤ Managing systems and networks

¤ Creating user interfaces, including navigation (browsing)

79
Issues to Consider for Intelligent
Agents
¨ Learning
¨ Performance

¨ Multiagents

¨ Cost justification

¨ Security and privacy

¨ Ethical issues

¨ Acceptance

80