Natural Language and Computer Programs

Natural Language and Computer Program Everyone who wishes to explain to the elderly the behavior of a computer or even a video recorder is familiar with why natural language manipulation is the goal of computer science researchers. In short, most people do not want to learn computer language to use their own electronic equipment. For people to use computer-based systems effectively, they must be programmed to understand natural languages ​​- the languages ​​spoken by ordinary people - and they respond in a physical way There must be.

It all started with ELIZA, an early natural language processing computer program created in 1964 (!) By Joseph Weizenbaum at the MIT Artificial Intelligence Laboratory using the LISP computer programming language (!) In 1964. By creating a surface for human-machine communication using conversation of pattern matching and word replacement simulation, ELIZA gives the user an illusion of understanding, but in reality there is no built-in framework for contextual response . All of these are done with a powerful cloud-based infrastructure with a powerful AI assistant "digital employee". Then you can teach customer support, selling insurance policies, providing advice on stock trading, listing the cars you want to purchase, and creating a continuing list. The more AI assistants you use, the better they will be. I do not outsource overseas service desks any more. The future of "robot technology" is actually software, not hardware.

ELIZA is a natural language processing computer program originally written by Massachusetts Institute of Technology Joseph Weizenbaum. The program is a simple therapist who interacts with people and demonstrates trivial and superficial communication between people and machines. This is implemented in the Fortran preprocessor. It mimics the Lisp function for combining list processing and nested expressions. Later written in Lisp by Bernie Cosell. MACSYMA (Symbolic Manipulator of MAC) is one of the computer algebra systems developed originally by Massachusetts Institute of Technology Project MAC. MACSYMA was written in the Lisp dialect called MacLisp. MacLisp was one of the biggest Lisp programs there. In 1982, MACSYMA was licensed to Symbolics, its computer hardware company whose main business is production machines and its architecture is optimized for the development and interpretation of the Lisp program.

Natural language processing (NLP) is a computer science, information engineering, and artificial intelligence, including computer programming methods for processing and analyzing interactions between computers and human (natural) languages, especially large amounts of natural language data Sub-area of. In the Georgetown experiment of 1954, we automatically translated over 60 Russian sentences into English. The authors claim that machine translation will be a solution within 3 to 5 years. However, the actual progress was much slower, and after it turned out that the study of 10 years did not meet expectations in the ALPAC report of 1975, the funds of machine translation were greatly reduced. Further study on machine translation until the late 1980s when the first statistical machine translation system was developed