STOCHASTIC GRADIENT BOOSTING MODEL-BASED CONTENT ANALYSIS IN ONLINE SOCIAL MEDIA

Abstract

This project describes content analysis of text with to identify suicidal tendencies and types. This article also describes how to make a sentence classifier that uses a neural network created using various libraries created for machine learning in the Python programming language. Attention is paid to the problem of teenage suicide and «groups of death» in social networks, the search for ways to stop the propaganda of suicide among minors. Analysis of existing information about so-called «groups of death» and its distribution on the Internet.

The study experience of content analysis of suicidal statements on the Internet of persons with different levels of suicidal activity» collects data from the pages of people who have committed suicide or are potential suicides. By analyzing the collected information, program called TextAnalyst explores the causes of suicidal behavior and their feelings. The aim of the current study is to classify sentences into suicidal and non-suicidal using a neural network. In our system, according to random text, it is necessary to determine whether it is suicidal or not, i.e., to solve the problem of its binary classification. Classification is the distribution of data by parameters.