You are on page 1of 9

1

A
PAPER
ON
TYPES OF EXPERT SYSTEMS
(MODEL-BASED, CASE-BASED AND FRAME-BASED EXPERT SYSTEMS)

Written By
FALOMO OLANIYI OLUWATOSIN
CSC/09/8180

COURSE TITLE: EXPERT SYSTEM







2

1.0 CASE BASED EXPERT SYSTEMS
Unlike rule-based systems which use facts and rules, Case-based expert systems use cases and
their entire solutions where cases represent problems.
In most cased based expert systems, a case is represented with a particular name, a set of facts,
and an outcome representing the result of the problem for decision, solution, or classification it
poses. In some systems, a case may also record an explanation of the outcome, the method of
solution, and the context in which the steps that led to the solution were taken. Each case is
indexed accordingly and linked to other cases directly or via the index.

The major elements in a case-based expert system are
The case-base set of cases,
The index library used to efficiently search and quickly retrieve cases that are most
appropriate or similar to the current problem.
Similarity metrics used to measure how similar the current problem is to the old cases
retrieved by searching the index library, and
Adaptation module This module uses either of the methods listed below;
Structural Adaptation : creates a solution for the current problem by either
modifying the solution
Derivational Adaptation : creates a new solution using the same process as was
used in the similar past case.
Case-based expert systems are also referred to as Case-Based Reasoning Systems (CBRS). Case-
based Reasoning is based on the concepts of analogy as well as similarity. Similarity is used to
find similar cases while analogy is used to find solutions for similar cases, in order to derive a
solution for the new case.
In case-based systems, the whole problem definition known as the case, is compared with similar
problems, and the entire solution is applied at once.
Operation Cycle of Case-based Reasoning Systems
The operation of a CBR system involves the four processes described below;
3

i. Retrieve the most similar case(s).
A case based system retrieves an old similar case in order to solve a new case. The
system uses two main methods to retrieve similar cases, they include;
Inductive Retrieval: Cases are selected based on the usefulness of the information
they carry as regarding the new case (problem) at hand.
Nearest-Neighbor : Cases are compared using some distance like Euclidean
distance.
||x-y||=( (x
1
-y
1
)
2
+..+(x
m
-y
m
)
2
)
Where x=(x
1
,..,x
m
) and y=(y
1
,..,y
m
) are two different cases.
ii. Reuse the case(s) to attempt to solve the problem at hand.
The solution of the retrieved case in the above step is used for the new case.
iii. Revise the proposed solution.
The solution of the retrieved case can be revised or adapted.
iv. Retain the new solution as a part of a new case.
If the proposed solution is revised and tends to solve the case at hand, the new case is
retained in the database together with the new solution.

Figure 1 Operation Cycle of Case based Expert system.

4

Advantages of Case-based Expert systems
1. Offers a cost-effective solution to the knowledge acquisition bottleneck problem
2. Case-based systems can learn from experience and so can be self-maintaining.
3. They are used when the problem is dynamic or not well-understood.
Disadvantages
1. Can take large processing time to find similar cases in the case-base.
2. Can take large storage space for all the cases.















5

2.0 FRAME BASED EXPERT SYSTEMS
Frame based expert systems use frames to encapsulate data and methods about a particular
entity.
Frames provide a natural way for the structured representation of knowledge. Frames are an
application of object-oriented programming for expert systems.
A frame is a data structure with knowledge about a particular object or concept (Negnevitsky,
2005). A frame provides a means of organizing knowledge in slots to describe various attributes
and characteristics of the object (entity).
Each frame has its own name and a set of attributes (slots) associated with it. For instance,
username, user_id, and password are slots in the frame User. Each attribute or slot in a frame
has a value assigned to it. Slots are used to store values. A slot may contain a default value or a
pointer to another frame, a set of rules or procedure by which the slot value is obtained.
Information included in a slot
Frame name.
Relationship of the frame to the other frames. .
Slot value. The value can be symbolic, numeric or Boolean. Slot values can be
assigned when the frame is created or during the operation of the expert system
Default slot value.
Range of the slot value
Procedural information. A slot can have a procedure attached to it, which is
executed if the slot value is changed or needed.
Frame-based expert systems support class inheritance, an object oriented programming concept.
The fundamental idea of inheritance is that attributes of the class-frame represent things that are
typically true for all objects in the class. However, slots in the instance-frames can be filled with
actual data uniquely specified for each instance.
6

Expert systems are required not only to store the knowledge but also to validate and manipulate
this knowledge. To add actions to frames, the knowledge engineer makes use of methods and
demons. Frame based expert systems use the following two methods;
WHEN_CHANGED and
WHEN_NEEDED
A WHEN_CHANGED method is invoked when the value of its attribute changes WHILE a
WHEN_NEEDED method is used to obtain the attribute value when its needed.
A WHEN_NEEDED method is invoked when information associated with an attribute is needed
for solving the problem, but the attribute value is undetermined
A method is a procedure associated with a particular frame slot that is executed whenever
requested while a demon has an IF-THEN structure and it is executed whenever an attribute in
the demons IF statement changes its value.
How does an inference engine work in a frame based system?
In all rule-based systems, where the inference engine links the rules present in the knowledge
base with data give in the database. In these RB systems, when the goal is set-up, the inference
engine searches the knowledge base to find a rule that has the goal in its consequent. If such a
rule is found and if its conditional part correlates with data in the database , the rule is fired and
the specified goal (the object ), obtains its value. If no rules are found that can generate a value
for the goal, the system queries the user to supply that value.
But in a frame-based system, rules play an auxiliary role. In these systems, Frames are used as
the major source of knowledge, and both methods and demons are used to add actions to the
frames.
Thus, the goal in a frame-based system can be established either in a method or in a demon.



7

3.0 MODEL BASED EXPERT SYSTEM
According to McGraw-Hill Science & Technology Dictionary, a model-based expert system is
an expert system that is based on knowledge of the structure and function of the object for which
the system is designed.
Many expert systems like case-based and frame-based expert systems rely on highly-compiled,
large-gained knowledge for their reasoning power. But model-based expert systems rely on an
alternative approach known as model-based reasoning, which uses a detailed model of the
objects in the domain and the operations that act on those objects.
Model based reasoning can make expert systems more robust by providing them with the ability
to do deep reasoning. This allows the system to react sensibly to new faults and provide better
explanations. The absence of the knowledge engineering bottleneck makes MBR systems easier
to develop.
The well-developed area in Mobile based reasoning deals with automating the process of
diagnosis. Given a system model and a set of observations or symptoms, diagnosis is the process
of identifying one or more faulty components that can account for the systems. The model-based
approach to diagnosis employs an explicit system model containing details of the components
and their behavior, and a description of the system structure. Using some given input states, a
simulation algorithm is used to generate system behavior that can be compared with the real
system. When a fault occurs, the system symptoms will conflict with the model output and the
discrepancies can be used to generate candidate components that may be responsible for the
faults.
When model-based reasoning is used, the program creates its solution anew for each problem,
based on its model of the domain, Model-based expert systems doesnt rely on pre-compiled
solutions. The expert system can easily solve any problem that is derivable from the model of the
domain.
Some model-based expert systems require complex control structure to reduce the amount of
search for possible solutions.
Model based reasoning is one of the knowledge-based methods used for fault detection.
8

The model- based system tells what to expect, and when observations differ from these
expectations, it will lead to identification of faults.
Thus, a knowledge based reasoned whose analysis is founded directly on the specification and
functionality of a physical system is called a MODEL-BASED system.
Qualitative model-based reasoning includes;
A description of each component in the device
Observation of the devices actual performance
A description of the internal structure of the device.
It has the following advantages and disadvantages as a method for fault detection;
Advantages of Model-based diagnostic expert systems
1. Knowledge acquisition is not needed.
2. Can deal with unexpected cases.
3. Flexibility in the cases of design changes.
4. Dynamic fault detection.
5. Ability of diagnosing incipient faults.
Disadvantages of Model-based diagnostic expert systems
1. Domain dependent.
2. Knowledge base is very demanding.
3. Difficulty in the isolation of faults
Example of Model based Expert Systems
1. GENEX II (Koton, 1985) :
The system uses model-based reasoning to solve problems about the behavior of bacterial
operons, a subfield of molecular biology.
The system takes as input a description of an operon and other information, then attempt
to get the biological control mechanism causing the observed behavior of the operon.
9

GENEX II can solve a greater variety of problems and complex problems unlike its older
version, GENEX which is Rule-Based.


REFERENCES
P.A Koton (1985), Empirical and Model-based reasoning in expert systems, Proceedings of the
International Joint Conference on Artificial Intelligence.
C.Angeli , Diagnostic Expert Systems: From Experts Knowledge to Real-Time Systems,
Advanced Knowledge Based Systems: Model, Applications & Research (Eds. Sajja & Akerkar),
Vol. 1, pp 50 73, 2010.
Bryan S. Todd (February 1992), An Introduction to expert Systems, Oxford University
Computing Laboratory Programming Research Group, England
Carol E. Brown et al, Introduction To Artificial Intelligence And Expert Systems.
Klaus-Dieter Althoff, Case-Based Reasonong and Expert Systems, Competence Center Case-
Based Reasoning German Research Center for Artificial Intelligence (DFKI) Intelligent
,Information Systems Lab Institute of Computer Science ,University of Hildesheim

You might also like