Abstract:
The article explores the possibility of using an associative-semantic preprocessor for special text processing in natural language. The use of associations allow to abstract from the direct meaning of a word and to replace it with a set of other words. This has also the opposite effect: by typing words (associations) a person is able to restore the search word, which allows to form a query in a natural language without knowing the keywords or terms of a particular domain but at the same time to receive the required result, in contrast to systems oriented to frequency occurrences of words. In the semantic processing of text using associations, the order of words and their number are not important, which allows a person to communicate with the machine without formulating phrases in a special way, since the interactive dialog system itself will process the request clearing everything else. The use of a special text preprocessor based on the associative-semantic processing of text allows interactive systems to be able to understand the topic of the machine's dialogue with the user, improve interaction by communicating in a natural language, and also to simplify the process of system creation and development.