AI Technique
Based on the underlying assumption that knowledge can be converted in to symbols, some techniques of solving AI problems are designed. So we can define AI technique as “It is a method that makes use of knowledge which is represented in a particular format to arrive at new results or new knowledge”. We can say that AI technique can work only if knowledge is represented in a particular way and it has following properties.
- The knowledge should be general i.e. When ever we talk about the solution to a given problem we must reach a general solution which can be applicable to other problems as well. If knowledge is represented separately for each situation then huge amount of memory will be required and which will be practically infeasible as cost of implementing such a solution will be extremely high. Knowledge without this characteristic will be called data rather than knowledge.
- It can be used in many situations even if it is not perfect or complete. e.g. Chess playing.
- It should be easy to modify.
- It should be able to overcome its own volume by reducing the range of overall possibilities (Combinational Explosion).
Types of AI Techniques:
There are following three AI techniques available.
- Use of Knowledge:
- Search:
- Generalization: