Knowledge Representation in AI examples
Knowledge Representation is a progression that starts with data that is of limited utility. Data, when processed, becomes information, information when interpreted or evaluated becomes knowledge, and understanding of the principles embodied with the knowledge is wisdom.
Properties of Knowledge Representation:
1. Representation Adequacy: It is defined as an ability to represent the required knowledge.
2. Inferential Adequacy: It is defined as an ability to manipulate the knowledge represented to produce new knowledge corresponding to that inferred from the original.
3. Inferential Efficiency: It is defined as an ability to direct the inferential mechanisms in the most productive directions by string-appropriate guides.
4. Acquisition Efficiency: It is defined as an ability to acquire new knowledge using automatic methods where possible rather than relying on human intervention.
Artificial Intelligence System must be capable of doing three things:
a. Store knowledge
b. Apply the knowledge stored to solve problems
c. Acquire new knowledge through experience.
Types of Knowledge:
There are mainly seven types of knowledge in Artificial Intelligence systems.
1. Inheritable Knowledge: We can inherit a certain type of knowledge also.
2. Inferential Knowledge: The knowledge representation method which can use an inference mechanism to use this knowledge is called Inferential Knowledge.
3. Relations knowledge: This knowledge is represented in relations/tables in DBMS.
4. Heuristic Knowledge: Heuristic means ‘rules of thumb’. Guessing is a way of making conclusions. It is a form of judgemental information.
5. Commonsense Knowledge: A knowledge gained by our experience about any general phenomenon is called Commonsense Knowledge.
6. Explicit Knowledge: A knowledge that an individual gains explicitly is called explicit knowledge.
7. Uncertain Knowledge: Today, we are surrounded by many uncertainties. These uncertainties need to be considered as they also provide some sort of language.