Types of Knowledge in Artificial Intelligence - Definitions | Artificial Intelligence Notes

Types of Knowledge in Artificial Intelligence

There are mainly five types of knowledge in artificial intelligence.In this article we will cover 5 main types of knowledge in AI.

Types of Knowledge in Artificial Intelligence - Definitions

1. Declarative Knowledge
2. Procedural Knowledge
3. Heuristic Knowledge
4. Meta-Knowledge
5. Structural Knowledge

While describing all of these types we will stay simple enough to be understandable.
Firstly lets define:

What is Knowledge in Artificial Intelligence?

We can define Knowledge as:

Understanding of a subject area, which can be used for further deductions.

Knowledge in Artificially intelligent systems can be from Knowledge base (KBS) which is feed in it by a Knowledge engineer or by Learning itself due to precepts from environment and solving problems.
Expert Systems in Artificial Intelligence have the ability to expand their knowledge and learn things. Learning enables them to do task from same population of data better and more efficient next time. This also increase their knowledge table and enhances their capabilities.
Now lets jump to explain types of knowledge in artificial intelligence.

Types of Knowledge in Artificial Intelligence - Definitions

1. Declarative Knowledge

The first types of knowledge in artificial intelligence, Declarative Knowledge in artificial intelligence is the type of knowledge that is based upon concepts, facts and objects. Declarative Knowledge provides necessary information about the problem that can either be true or be false.
For example a Smart System Declares that : It will rain today.
This is a declarative statement which can either be true if it actually rains or be false if it doesn't.

2. Procedural Knowledge:

The first types of knowledge in artificial intelligence, Procedural Knowledge provides information on basis of rules, strategies, agendas and procedures. Procedural Language mainly directs on how to do something.
For example: Imagine a Form Filling System(Software), Which will give you step by step instruction on how to fill a form. What details you should put in. These instructions will be the procedures that we are following thus, falls in category of procedural language.

The conventional protocol is a bike driving. You likely tried to understand it until a few moments if somebody was showing you how to ride a bike, no matter what they said. It soon became explicit wisdom when you found it out. That is, the sort of information that is difficult to understand because it is placed in your mind subconsciously.

3. Heuristic Knowledge:

Heuristic Knowledge is based upon thumb rule. Thumb rule can be defined as:
"Principle with Broad applications that is not intended to be strictly reliable for every situation."
Heuristics Knowledge is based more upon Practical experiences rather then theoretical knowledge.

Often referred to merely as heuristic, any issue fixing or self finding strategy involving an operational technique is adequate to reach an instant objective, not expected to be optimum, ideal, logical, or reasonable. When it is not possible or practicable to find an ideal answer, heuristic techniques can be employed to speed up the method of discovering a suitable alternative.

4. Meta-Knowledge:

It is the Knowledge about knowledge. It is helpful in enhancing efficiency of a given problem. Meta Knowledge provides guidance about how to get information about some knowledge as well.

Knowledge not specific to the domain, but rather the inner composition of the field. Meta-knowledge is used by several previous schemes to enhance the complexity of the user environment, retain the understanding basis and regulate the inference engine.

5. Structural Knowledge:

Structural knowledge in artificial intelligence is the information based on rules, concepts and relationships. It helps us by providing information for developing structures and overall mental model of the desired system.