5.4 Other Specialized Information Systems

5.4.1 An overview of Artificial Intelligence

Artificial Intelligence (AI) is a field of science and technology based on disciplines such
as computer science, biology, psychology, linguistic, mathematics and engineering. The
goal of AI is to develop computers that can think, see, hear, walk, talk and feel.

A major power of AI is the development of computer functions normally associated with
human intelligence such as reasoning, learning and problem solving.

Artificial Intelligence systems include the people, procedures, hardware, software data
and knowledge needed to develop computer systems and machines that demonstrate
the characteristics of intelligence. Characteristics of intelligence include the following;
Learn from experience and apply the knowledge acquired from experience.

Handle complex situations
Solve problems when important information is missing.
Determine what is truly important
React quickly and correctly to a new situation
Understand visual images
Process and manipulate symbols
Be creative and imaginative
Use heuristics (learn by discovering) or guesses

One of the problems in AI is arriving at a working definition of real intelligence against
which to compare the performance of an artificial intelligence system (table x).



5.4.1.1 Major Application Areas of Artificial Intelligence

AI applications can be grouped under three major areas. They are Cognitive science,
robotics and natural interfaces.



Cognitive science:

This area of AI is based on research on biology, neurology, psychology, mathematics
etc.

It focuses on researching how the human brain works and how humans think and learn.
The results of such research in human information processing are the basis for the
development of a variety of computer-based applications in AI. Applications in the
cognitive science area of AI include the following;

• Development of expert systems and other knowledge based systems that add a
knowledge base and some reasoning capability to information systems.
• Adaptive learning systems that can modify their behaviors based on information
they acquire as they operate. Example: Chess playing systems
• Fuzzy logic systems can process data that are incomplete or ambiguous, that is
fuzzy data (example: data termed as low, very high, reasonable). They work as
humans by developing approximate inferences and answers to solve
unstructured problems with incomplete knowledge.
• Genetic algorithms software uses mathematics functions to stimulate evolutionary
processes that can generate increasingly better solutions to problems.
• Intelligent agents use expert systems and other AI technologies to serve as
software surrogates for a variety of end user applications.

Robotics
Basic disciplines of robotics include AI, Engineering and physiology. This technology
produces robot machines with computer intelligence and computer controlled, human
like physical capabilities.

This area includes applications designed to give robots the powers of

Visual perception: sight
Tactility: ability use the sense of touch
Dexterity: ability in using hands skillfully
Locomotion: physical ability to move over any surface
Navigation: the intelligence to properly find ways to a destination.

Many applications of robotics exist and research into these unique devices continue.

Robots are used to assemble and paint products in manufacturing. In military
applications robots are moving beyond movie plots to become real weapons.

Natural interfaces

Development of natural interfaces is essential to the natural use of computers by
humans. This area of applications are designed for,
Natural languages and Speech recognition: ability to communicate with computers and
robots in conversational human languages and have them understand us as
easily as we understand each other. This involves research and development in
linguistics, psychology, computer science and other disciplines.

Multisensory Interfaces: development of multisensory devices that use variety body
movements to operate computers.

Virtual reality: using multisensory human computer interfaces that enable human users
to experience computer simulated objects, spaces, activities and worlds as they actually
exist.

Neural networks

These are computing systems modeled based on brain’s mesh-like network of neurons.
Although neural networks are simpler than the human brain in architecture, similar to
brain, the interconnected processors in a neural network operate in parallel and interact
dynamically with each other. This enables neural networks to process many pieces of
data at once to learn to recognize patterns.

Some of the specific features of neural networks include the following:
• The ability to retrieve information even if some of the neural nodes fail
• Fast modification of stored data as a result of new information
• The ability to discover relationships and trends in large databases
• The ability to solve complex problems for which all the information is not present
An Example of the use of neural networks:
A neural network can be trained to learn which characteristics result in good or
bad loans.

5.4.2 Expert Systems

5.4.2.1 An Overview of a Expert System

An expert system is a knowledge-based information system that uses its knowledge
about a specific, complex application area to act as an expert consultant to end users.
Expert systems provide answers to questions in a very specific problem area by making
humanlike inferences about knowledge contained in a specialized knowledge base. They
must also be able to explain their reasoning process and conclusions to a user.

Expert systems can provide decision support to end users in the form of advice from an
expert consultant in a specific problem area.

Knowledge-based information system (KBIS)

A KBIS adds a knowledge base to the major components found in other types of
computer based information systems.An Expert system consists of a collection of integrated and related components including a knowledge base, an inference engine, an explanation facility, a knowledge base acquisition facility and a user interface.



The user interacts with the user interface which interacts with the inference engine. The
inference engine interacts with the other expert system components. These components
must work together to provide expertise.

Knowledge base

The knowledge base stores all relevant information data, rules, cases and relationships
used by the expert system.

A knowledge base is a natural extension of a database and an information and decision
support system. With a knowledge base, we try to understand patterns and relationships
in data as a human expert does in making intelligent decisions.

A knowledge base must be developed for each unique application. It can include generic
knowledge from generic theories that have been established over time and specific
knowledge that comes from more recent experiences and rules of thumb. Knowledge
bases however go far beyond simple facts, also storing relationships, rules or frames
and cases.

Rules suggest certain conclusions, based on a set of given facts. Normally rules are
stored as if-then statements.

Inference engine

The inference engine processes the rules, data, and relationships stored in the
knowledge base to provide answers, predictions, and suggestions the way a human
expert would. Two common methods for processing include backward and forward
chaining. Backward chaining starts with a conclusion, then searches for facts to support
it; forward chaining starts with a fact, then searches for a conclusion to support it. Mixed
chaining is a combination of backward and forward chaining.

Explanation facility

Component of an expert system that allows a user or decision maker, to understand how
the expert system arrive at a certain conclusion or a result.
The expert system using the explanation facility can indicate all the facts and rules that
were used in reaching the conclusions.

Knowledge base acquisition facility

This acts an interface between experts and the knowledge base. The overall purpose of
the knowledge acquisition facility is to provide a convenient and efficient means for
capturing and storing all components of the knowledge base. Knowledge acquisition
software can present users and decision makers with easy to use menus. After filling the
appropriate attributes, the knowledge acquisition facility correctly stores information and
relationships in the knowledge base.

Regardless of how the knowledge is acquired, it is important to validate and update the
knowledge base frequently to make sure that it is still accurate.

The knowledge acquisition facility acts an as interface between experts and the
knowledgebase.

User interface

Specialized user interface software is employed for designing, creating, updating and
using expert systems. The main purpose of the user interface is to make the
development and use of an expert system easier for users and decision makers.

Expert systems place more emphasis on directing user activities than other types of
systems. Text oriented user interfaces (using menus, forms and scripts) are more
common in expert systems than the graphical interfaces often used with DSSs.

5.4.2.2 Applications of an Expert System

Using an expert system involves an interactive computer based session in which the
solution to a problem is explored, with the expert system acting as a consultant to an end
user. The expert systems asks questions from the user, searches its knowledge base for
facts and rules or other knowledge, explains its reasoning process when asked and
gives expert advice to the user in the subject area being explored.

Expert systems are used in many different fields, including medicine, engineering,
physical sciences and business.

Now expert systems help to diagnose illnesses, search for minerals, analyze
compounds, recommend repairs, and for financial planning.

• Credit granting and loan analysis: In banks, Expert systems are used to review an
individual’s credit application and credit history data to make a decision on
whether to grant a loan or approve a transaction.
• Stock picking: Some expert systems are used to help investment professional
pick stocks and other investments
• Catching cheats and terrorists: Some gambling casinos use expert system
software to catch gambling cheats.
• Budgeting: Automotive companies use expert systems to help budget, plan and
coordinate prototype testing programs
• Games: Some expert systems are used for entertainment- crossword puzzles
• Information management and retrieval: Expert system agents help managers find
the right data and information while filtering out irrelevant facts that might delay
timely decision making.
• Manufacturing: Expert systems can be used to spot defective welds during the
manufacturing process. The expert system analyzes radiographic images and
suggests which welds could be damaged.

5.4.2.3 Benefits and Limitations of an Expert System

Benefits

An expert system captures the expertise of an expert or group of experts in a computerbased
information system. Thus it can outperform a single human expert in many
problem situations. That’s because an expert system is faster and more consistent, can
have the knowledge of several experts, and does not get tired or distracted by overwork
or stress.

Expert systems also help preserve and reproduce the knowledge of experts. They allow
a company to preserve the expertise of an expert before he/she leaves the organization.
This expertise can then be shared by reproducing the software and knowledge base of
the expert system.

Expert systems can explain how and why a decision or solution was reached. Ability to
explain its reasoning process can be the most valuable feature of a computerized
system.

ESs can display intelligent behavior: that is proposing new ideas or approaches to
problem solving by considering a collection of data.

ESs can evaluate complex relationships to reach conclusions and solve problems.
ESs can deal with uncertainty: that is the ability to deal with incomplete or not completely
accurate knowledge.

Limitations

Expert systems excel only in solving specific types of problems in a limited domain of
knowledge. They are poor in solving problems requiring a broad knowledge base and
subjective problem solving. For an ES the primary source of knowledge is a human
expert. If this knowledge is incomplete or incorrect it will lead to errors in the system.
Other development errors involve poor programming practices.

Major limitations of expert systems arise from their limited focus, inability to learn,
maintenance problems and development cost. Example for limited focus: An ES
designed to provide advice on how to repair a machine is unable to decide when or
whether to repair it.

They do well in operational or analytical tasks, but can make mistakes at subjective
managerial decision making.

Expert systems cannot refine its own knowledge. A programmer must provide
instructions to the system that determine how the system is to learn from experience.
Expert systems are difficult and costly to develop and maintain properly. The cost of
knowledge engineers, lost expert time, and hardware and software resources may be too
high.

ESs are not used in a large number of organizations. That means they have not been
widely tested in corporate settings. Some ESs are difficult to control and use. Therefore,
users require the support of trained computer personnel to use the system.

Expert systems cannot easily handle knowledge that has a mixed representation.
Knowledge can be represented through defined rules, in comparison with similar cases,
and in various other ways. An ES in one application might not be able to deal with
knowledge that combines both rules and cases.

ESs are difficult to maintain. Some are not responsive or adaptive to changing
conditions. Adding new knowledge and changing complex relationships may require
sophisticated programming skills. Expert systems can be expensive to develop when
using traditional programming languages and approaches. Development cost can be
greatly reduced through the use of software for expert system development.

Expert systems raise legal and ethical concerns too. People who make decisions and
take action are legally and ethically responsible for their behavior. When expert systems
are used to make decisions or help in the decision-making process, who is legally and
ethically responsible? These legal and ethical issues have not been completely resolved.