Linguistic Data Model for Natural Languages and Artificial Intelligence. Part 3. Recognition

Abstract

Introduction. The paper continues a series of publications on relations linguistics (hereinafter R-linguistics) and is devoted to the analysis of the recognition problem in relation to the approach considered in the series. Recognition directly affects language forms, especially since the model used in the framework of R-linguistics creates significant features in recognition.

Methodology and sources. The research methods consist in the development of the necessary mathematical concepts for linguistics in the field of identification, which uses the verbal approach to previously obtained results on identification in linguistic spaces.

Results and discussion. As a recognition problem in R-linguistics, two tasks are identified: types recognition and signs value recognition. Each of these tasks has a specific dimension, the extension of tuples of parameters, blocking errors in recognition, etc. In addition, the presence of a linguistic model helps to simplify the solution of these problems. In the section the features and ways of solving both problems of recognition are formulated taking into account the stated specifics.

In this section, based on the material of all three parts, the general contours and properties of the linguistic model of the world are described. It also discusses various aspects of recognition associated with linguistic spaces: variables, memory, expansion problems, etc.

Conclusion. In conclusion, the law of creative thinking is formulated, which follows from a linguistic data model.