Home | | Artificial Intelligence | Issues in Knowledge Representation

Chapter: Artificial Intelligence(AI) : Knowledge Inference

Issues in Knowledge Representation

The fundamental goal of Knowledge Representation is to facilitate inferencing (conclusions) from knowledge.

Issues in Knowledge Representation

 

The fundamental goal of Knowledge Representation is to facilitate inferencing (conclusions) from knowledge.

 

The issues that arise while using KR techniques are many. Some of these are explained below.

 

Important Attributes :

 

Any attribute of objects so basic that they occur in almost every problem domain ?

 

Relationship among attributes:

 

Any important relationship that exists among object attributes ?

 

Choosing Granularity :

 

At what level of detail should the knowledge be represented ?

 

Set of objects :

 

How sets of objects be represented ?

 

Finding Right structure :

 

Given a large amount of knowledge stored, how can relevant parts be accessed ?

Important Attributes :

 

 

There are attributes that are of general significance.

 

There are two attributes "instance" and "isa", that are of general importance. These attributes are important because they support property inheritance.

 

Relationship among Attributes : (Ref. Example- Fig. Inheritable KR)

 

The attributes to describe objects are themselves entities they represent.

 

The relationship between the attributes of an object, independent of specific knowledge they encode, may hold properties like:

 

Inverses, existence in an isa hiera hy, techniques for reasoning about values and single valued attributes.

 

Inverses :

 

This is about consistency check, while a value is added to one attribute. The entities are related to each other in many different ways. The figure shows attributes (isa, instance, and team), each with a directed arrow, originating at the object being described and terminating either at the object or its value.

 

There are two ways of realizing this:

 

first, represent two relationships in a single representation; e.g., a logical representation, team(Pee-Wee-Reese, Brooklyn–Dodgers), that can be interpreted as a statement about Pee-Wee-Reese or Brooklyn–Dodger.

 

second, use attributes that focus on a single entity but use them in

 

pairs, one the inverse of the other; for e.g., one,

 

Dodgers , and the other, team = Pee-Wee-Reese, . . . .

This second approach is followed in semantic net and frame-based systems, accompanied by a knowledge acquisition tool that guarantees the consistency of inverse slot by checking, each time a value is added to one attribute then the corresponding value is added to the inverse.

◊ Existence in an "isa" hiera hy :

 

This is about generalization-specialization, like, classes of objects and specialized subsets of those classes. There are attributes and specialization of attributes.

 

Example: the attribute "height" is a specialization of general attribute

 

"physical-size" which is, in turn, a specialization of "physical-attribute". These generalization-specialization relationships for attributes are important because they support inheritance.

 

Techniques for reasoning about values :

 

This is about reasoning values of attributes not given explicitly. Several kinds of information are used in reasoning, like,

 

height : must be in a unit of length,

 

age : of person can not be greater than the age of person's parents.

 

The values are often specified when a knowledge base is created.

 

Single valued attributes :

 

This is about a specific attribute that is guaranteed to take a unique value.

Example : A baseball player can at time have only a single height and be a member of only one team. KR systems take different approaches to provide support for single valued attributes.

Choosing Granularity

 

What level should the knowledge be represented and what are the primitives ?

 

 Should there be a small number or should there be a large number of low-level primitives or High-level facts.

 

 High-level facts may not be adequate for inference while Low-level primitives may require a lot of storage.

 

Example of Granularity :

 

Suppose we are interested in following facts

 

John spotted Sue.

 

This could be represented as

 

Spotted (agent(John), object (Sue))

 

Such a representation would make it easy to answer questions such are

 

Who spotted Sue ?

 

Suppose we want to know

 

Did John see Sue ?

 

Given only one fact, we cannot discover that answer.

 

We can add other facts, such as

 

Spotted (x , y)                    saw (x , y)

We can now infer the answer to the question.

 

Set of Objects KR - issues

 

Certain  properties  of  objects  that  are  true  as member of  a set  but not as  individual;

Example : Consider the assertion made in the sentences

"there are more sheep than people in Australia", and "English speakers can be found all over the world."

 

To describe these facts, the only way is to attach assertion to the sets representing people, sheep, and English.

 

The  reason to represent sets of objects is :

 

If a property is true for all or most elements of a set, then it is more efficient to associate it once with the set

rather than to associate it explicitly with every elements of the set . This is done in different ways :

 

in logical representation through the use of universal quantifier,  and

 

in hiera hical structure where node represent sets, the inheritance propagate set level assertion down to individual.

 

Example: assert large (elephant);

 

Remember to make clear distinction between,

 

whether we are asserting some property of the set itself,

 

means, the set of elephants is large,         or

 

asserting some property that holds for individual elements of the set , means, any thing that is an elephant is large.

 

There are three ways in which sets may be represented :

 

Name, as in the example – Ref Fig. Inheritable KR, the node - Baseball-Player and the predicates as Ball and Batter in logical representation.

 

Extensional definition is to list the numbers,  and

 

In tensional definition is to provide a rule, that returns true or

 

false depending on whether the object is in the set or not.

 

 

Finding Right Structure

 

 

Access to right structure for describing a particular situation.

 

It requires, selecting an initial structure and then revising the choice. While doing so, it is necessary to solve following problems :

 

how to perform an initial selection of the most appropriate structure.

 

how to fill in appropriate details from the current situations.

 

how to find a better structure if the one chosen initially turns out not to be appropriate.

what to do if none of the available structures is appropriate.

 

when to create and remember a new structure.

 

There is no good, general purpose method for solving all these problems. Some knowledge representation techniques solve some of them.


Study Material, Lecturing Notes, Assignment, Reference, Wiki description explanation, brief detail
Artificial Intelligence(AI) : Knowledge Inference : Issues in Knowledge Representation |


Privacy Policy, Terms and Conditions, DMCA Policy and Compliant

Copyright © 2018-2024 BrainKart.com; All Rights Reserved. Developed by Therithal info, Chennai.