TWO-STAGE CLASSIFICATION METHOD

Teterin Nikolay Nikolaevich, Smolentseva Tatyana Evgenievna

The article discusses a two-level classification method aimed at improving the quality of reliability of detecting destructive user behavior in social networks. The relevance of the study is due to the need to automate content moderation processes on the Internet, which will allow us to quickly respond to potentially destructive behavior. The purpose of this work is to develop a two-level classification method designed to detect destructive user behavior in social networks. The proposed approach includes two stages: at the first level, binary classification is performed to filter data, and at the second level, multi-class classification is used to accurately determine categories. The key quality criteria for the two-level classification method are: adaptability to new forms of threats, interpretability of decisions made, and compliance with expert assessment.

Keywords:two-stage classification, detection, destructive behavior, social networks, heterogeneous data.