What Is Artificial Intelligence Programing?

Artificial intelligence programming language is a type of computer programming language with symbol processing and logical reasoning capabilities adapted to the fields of artificial intelligence and knowledge engineering. It can be used to write programs to solve non-numerical calculations, knowledge processing, reasoning, planning, decision making and other intelligent and complex problems. ^[1]

In fact, there are many artificial intelligence programming languages that correspond to a variety of different representations of knowledge. According to the corresponding knowledge representation is different. It can be divided into the following categories:

1. Language corresponding to knowledge representation of production rules . For example, the OPS (official production system) developed by CL Forgy, Carnegie Menon University, and others in 1977, at that time 'was used to

Artificial intelligence programming languages have a common feature, that is, these languages are designed independently for the problem to be solved, combined with knowledge representation, and completely separated from the characteristics of the contemporary computer's Neumann structure; they are in a higher level than process-oriented advanced programming Higher levels of abstraction in language. Therefore, programs written in these languages are often inefficient in modern computer environments, whether they are interpreted or compiled. Especially when the program is large in scale and complicated, it will waste a lot of system resources (mainly referring to the time occupied by the processor and the storage space), which will cause the system performance to be intolerable. ^[2]

The first is to develop a so-called next-generation computer that is fully adapted to a certain language. Such as LlSP machine, data stream machine, PROLOG machine, object-oriented architecture and so on. But practice has already shown that the prospect of this road is slim. This is because modern general-purpose computers with Neumann machines as the core have been widely popularized and their performance is continuously improved, while huge software resources have been accumulated. Any special machine that is not compatible with modern computers can at most play a role in certain occasions because it can meet individual special needs, and it is difficult to compete with modern general-purpose computers.

The second way is to combine several different styles of programming languages and develop compound or embedded languages in order to learn from each other's strengths and improve system performance. At present, the most popular is to integrate the design ideas of object-oriented languages into commonly used high-level languages. The C ++ language is a prominent example.

The third way is to make full use of the characteristics of the problem-oriented artificial intelligence programming language, and first select a certain language to write a concise and easy to debug program prototype. After verification, debugging, and then imitating this prototype, it is adapted into a process-oriented high-level language program, such as C or C ++, or even BASIC, to achieve the purpose of improving the final application system development quality and execution efficiency. There are many precedents for developing expert system prototypes using PROLOG, LISP, OPS, etc. ^[2]