USE OF TECHNOLOGY TEXTMINING FOR IDENTIFYING HIDDEN THREATS IN COMMUNICATIONS DISTRIBUTED BY SOCIAL NETWORKS
A solution to the problem of text analysis using TextMining technology to detect threats hidden in messages exchanged between users in social networks has been proposed. The possibilities of TextMining technology in the tasks of knowledge detection in unstructured information arrays are discussed. The sequence of text analysis is presented in the form of a methodology. The content of the stages of the methodology is disclosed and the main techniques used at each stage are considered. The methods of calculating the weighting functions are selected, on the basis of which a list of keywords and phrases is formed. Ways of building semantic networks based on a set of keywords are considered. To automate text analysis, a software package that implements TextMining technology has been developed. The functions of the software package include the identification of keywords, relationships, and the emotional portrait of the user, which allows you to move from data to their meaning and draw conclusions about the information security of the text.
Authors: B. Ya. Sovetov, T. M. Tatarnikova, A. I. Yashin
Direction: Informatics, Computer Technologies And Control
Keywords: TextMining, semantic network, keywords, word occurrence frequency, text meaning, theme, text classification, dictionary, text tonality, software package
View full article