AUTOMATIC CLASSIFICATION SYSTEM DOCSORTER
Purpose of the development:
The system is designed to find signs of a thematic or emotional class of text.
Recommended application field:
News agencies and analytical centers.
Technical characteristic:
The operating system is Windows XP or later.
Advantages over analogues:
Compared to other systems, DocSorter is focused on working with English, Ukrainian and Russian languages. Thanks to the use of semantic analysis and, in particular, the lexical and semantic base of UkrWordNet, the system focuses on the content proximity between text elements, and not on the immediate text dictionary.
The development stage readiness:
Ready for application
Description of the development:
() The system determines the elements of the text that correspond to a specific concept. This allows to determine the thematic elements available in the text, and with their help - the theme of the texts. The system learns from examples of texts which are collected in thematic catalogs, so it can evaluate diverse topics, such as physics, mathematics, politics, sports, etc. When using sets of texts that correspond to emotionally colored circumstances, for example, reviews of films, the algorithm can be used to determine the emotional coloring of the text.
Information about newness of the development:
there are Ukrainian patents -- 1 items
corresponds technical description
Ready for implementation
Possibility of transfer abroad:
Combinated reduction to industrial level Joint production, sale, exploitation
Photo
Country
Ukraine
For additional information turn to: E-mail: gal@uintei.kiev.ua
|