Types of Knowledge Representation in AI (Relational, Inheritable, Inferential, Procedural)

There are 4 types of knowledge representation in AI. They are:

1. Relational Knowledge
2. Inheritable Knowledge
3. Inferential Knowledge
4. Procedural Knowledge

All types of knowledge are discussed below.

Relational Knowledge in AI

It is also called simple relational knowledge. Relational knowledge in AI is the simplest way of storing facts. In relational knowledge each fact about a set of objects is set in columns systematically. Relational representation representation is popular in database systems.

Characteristics of Relational Knowledge Representation

• Storing facts is very simple.
• Each fact about a set of objects is represented in some columns.
• Low opportunity for inference.
• Knowledge basis for inference engines.
• Popular in DBMS where the relationship between different entities is represented.

Example of Relational Knowledge Representation

Developer	Language	Experience
Danny	PHP	5 years
Hans	Javascript	7 years
Louise	Java	4 years

Inheritable Knowledge in AI

Inheritable knowledge in AI representation shows a way to represent data where there is some hierarchy. Data can be inherited from the super classes in this representation. The super class holds all the data in the sub classes.

Characteristics of Inheritable Knowledge Representation

• All data is stored in a hierarchy of classes.
• All classes should be arranged in a common form or in hierarchy.
• Inheritance is properly applied in this representation.
• Elements inherit values from other members of a class.
• Inheritable knowledge shows a relation between instance and class called instance relation.
• Every individual frame can represent the collection of attributes and its value.
• Objects and values are represented in Boxed nodes.
• To point from object to values arrows are used.

Example of Inheritable Knowledge Representation

Algorithm to retrieve a value of an instance object in Inheritable Knowledge

1. Go to the object in the knowledge base.
2. If there is a value for the attribute show it.
3. Else look for a value of instance if not fail.
4. Else go to the node and find a value for the attribute and then show it.
5. Else search until a value is found for the attribute.

Inferential Knowledge in AI

Inferential knowledge in AI defines the knowledge as a formal logic condition and has a strict rule. The knowledge is generated from objects by studying their relation.

Characteristics of Inferential Knowledge Representation

• A set of strict rules is defined which can derive more facts, verify new statements and ensure correctness.
• Many inference procedures are available to in implement standard rules of logic.
• Popular in Artificial Intelligence sector such as automated theorem proving.
• Represents knowledge in the form of formal logics.

Example of Inferential Knowledge Representation

Consider below two statements in the figure:

Procedural Knowledge in AI

Procedural knowledge in AI represents control information which uses the embedded in the knowledge itself. This approach offers representing and using domain knowledge in AI. They are represented as small programs of how to proceed and perform specific operations. Procedural knowledge may include inferential efficiency but they do not have inferential adequacy or acquisition efficiency.

Characteristics of Procedural Knowledge Representation

• Procedural knowledge uses small programs and codes to describe how to proceed and to do specific things.
• Most important rule used is If-Then rule.
• Various coding languages is used such as LISP language and Prolog language.
• Representing and using domain knowledge in AI is easy in procedural knowledge.
• Extended logical inferences such as default reasoning facilities are available.
• All cases may not be represented in procedural knowledge.