What are the areas of artificial intelligence

Artificial Intelligence (AI)

Artificial Intelligence. 1. Term: Research into "intelligent" problem-solving behavior and the creation of "intelligent" computer systems. Artificial intelligence (AI) deals with methods that enable a computer to solve such tasks that, if solved by humans, require intelligence.

2. Sub-areas: The figure “Artificial Intelligence” shows a possible structure of Artificial Intelligence (AI).

A distinction is made between methods and applications of Artificial Intelligence (AI); important interdisciplinary connections are highlighted by dashed lines.

1. The most significant Method areas Artificial intelligence (AI) is the representation of knowledge as well as the inference and inferring the use of the represented knowledge.

2. Special requirements The linguistic means of expression in the creation of Artificial Intelligence (AI) programs, especially the need for symbol processing, make special (AI) programming languages ​​necessary. These provide, among other things, certain forms of knowledge representation and offer options for evaluating the knowledge, e.g. through built-in inference methods. A well-known example is the programming language Prolog (logical programming).

3. The field of application deals with the development of “automatic proofers” for mathematical theorems Inference systems. In addition, inference systems are also developed with the aim of expanding the query options for database systems based on the relation model, e.g. to include recursive database queries.

4. Closely related to systems of deduction is the area of automatic programming. On the basis of a formal specification, the program verification can be carried out automatically with the aid of an inference system. In addition, automatic programming also includes the automatic creation of executable programs from formal specifications as well as proof of correctness for hardware components (e.g. integrated circuits, hardware).

5. In the methods of understanding natural language and their application in the context of Language processing the results of linguistics are used, e.g. from syntax theory. In addition to speech analysis, speech recognition is an important task within this application area.

6. Computer vision and robotics deal, among other things, with the interpretation of data from the real physical environment.

a) Computer vision deals with the areas of image understanding (gray value analysis, etc.), scene analysis (e.g. recognition of geometric objects from line drawings) and shape perception (description of the meaning of a scene, e.g. through the construction of a semantic network).

b) For object recognition, computer vision in the robotics resorted to. In this classic area of ​​application, the planning and control of robot actions plays an essential role.

7. In the center of the method areas learning and Cognitive models are peculiarities of human intelligence.

a) An important goal of the area Cognitive models is the creation of computer programs that simulate human problem-solving behavior.

b) Subject of the Learning are methods which should enable computer programs not only to act on the basis of the already existing, represented knowledge, but to automatically expand knowledge by evaluating known problems and their solutions.

8. While learning should transfer human ability to learn to the computer, within the scope of the area of ​​application ICAI (Intelligent Computer Aided Instruction) tries to help people in the process of learning. In doing so, use is made of knowledge from pedagogy.

9. Heuristic search is a method area from the first beginnings of Artificial Intelligence (AI). One problem with developing game programs is finding “good” moves; Because of the combinatorial diversity, the number of possible moves explodes very quickly. With the help of heuristics, the search areas are narrowed down so that game situations can be analyzed faster and better.