What is Knowledge Representation in AI?
Knowledge representation in AI means how AI agents think and how thinking contributes to behavioral intelligence of the AI agent.
The knowledge from natural language of human is not understandable for a machine without a proper representation. Knowledge representation in AI offers to represent information about the real world so that a computer can learn in easily. It is very important for a computer to understand knowledge to utilize the knowledge to solve the complex real world problems such as communicating with human or diagnosing a medical report.
Knowledge Representation in AI (KR in AI) is not just storing data into database. It also ensures that the machine is learning from the knowledge and experiences so they can behave like a real human.
Elements in Knowledge Representation
Following are the elements of knowledge representation which needs to be represented in Knowledge Representation AI systems:
- Object: All the facts about objects in the real world. For example, ships.
- Events: Real world actions are known as events. For example, picnic.
- Performance: Performance means the accuracy of knowledge about anything. For example, performance of a engine.
- Meta-knowledge: Knowledge about the information we know is called meta-knowledge. For example memory data.
- Facts: Facts are the ever truth events in the real world. For example, sun rises at east.
- Knowledge-Base: The central collection of of knowledge is known as knowledge-base. Such as, knowledge base of a software.
- Knowledge: Knowledge is to know the information about facts, data, and situations.
How to Identify Good Knowledge Representation System?
A good knowledge representation system must have the following properties:
- Representational Accuracy: Knowledge Representation system should be able to represent all kind of knowledges.
- Inferential Adequacy: Knowledge Representation system should be able to manipulate the representational structures to form new knowledge from the existing structure.
- Inferential Efficiency: Knowledge Representation system needs to be efficient to direct the inferential knowledge mechanism into the most productive directions.
- Acquisition efficiency: KR system should be able to acquire the new knowledge easily using automatic methods.