Content Analysis of Extracted Suicide Texts From Social Media Networks by Using Natural Language Processing and Machine Learning Techniques

Abstract

This work is about the fact that using new technologies, and in particular Natural Language Processing techniques with implementations of Machine Learning models, humanity can be ready to save the souls of people. With the development of the Internet and its accessibility, young people devote more and more time to social networks. It is often easier for them to write about their mood online than to talk about their problem with specialists. By analyzing posts on social networks, the nature of the topic, the manner of writing and other details, it can draw many ideas about a person's mood, predict his or her subsequent actions and prevent a possible crime. Kazakhstan ranks third in terms of the number of suicides in the world. To prevent suicide, it was decided to use sentiment analysis as a technique in Natural Language Processing to identify the positive or negative implication. In order to create a database, it is necessary to collect data from social networks popular in Kazakhstan. The main goal of the work is to combine NLP and ML techniques to build suitable models to get the needed information.