Using automatic text categorization technologies in the modern educational process

Abstract

The active improvement of natural text processing gives a lot of opportunities for the development ofeducational technologies. It is well-known that the participants in the modern educational process are facedwith the need to quickly view and classify large volumes of text documents. This problem arises everywhere:while searching the Internet for information in digital libraries, working with text databases and other elementsof the educational process. The improvement of text categorization technologies can make information retrievalmore efficient and therefore help the learner to more quickly and efficiently find the necessary information.The research deals with the problem of text categorization by the example of its assignment to aparticular age audience. In the first place, the solution to this problem gives the ability to improve the relevanceof information retrieval and also allows improvement of the mechanisms of excluding unwanted requests fromthe search results (such as websites whose content is designed for another age category).The authors of the research are developing approaches to mathematical modeling of the textcategorization task. These approaches are implemented within a prototype software system for automatic textcategorization based on the age of the text audience. In developing this system the authors are using texts inRussian, but the proposed methods are universal and can be applied to other related tasks.