Застосування технології ml.net задля автоматичного сортування та тегування текстових документів для семантичної цифрової бібліотеки.

Abstract

The paper proposes to develop a tool that will allow creating a semantic network of files in addition to the standard hierarchical Windows file system, which could automatically sort text files for a semantic digital library by categories based on their content. The preparation of a corpus of texts for the purpose of automatically obtaining categories, tags and attributes to fill a semantic digital library is an extremely important and relevant technology. In order to implement such an add-on to the Semantic File System, it is proposed to apply ML.NET technology, this will allow easy integration with other ML.NET components or external libraries, expanding the possibilities of sorting text documents.