Problematic Use of the Internet - Using Machine Learning in a Prevention Programme

Abstract

The aim of the study is to use machine learning methods in building a prevention program for Problematic Use of the Internet. On the basis of the collected database, a multi-stage analysis was performed to construct a model for selecting the users of the prevention program in order to best match the content that will be included in it. The results were based on three decision tree models. The use of such a matching mechanism in a prevention programme is likely to increase its effectiveness and reach a wider audience. Additionally, the problematic use of the Internet itself is a heterogeneous phenomenon and proper identification of the most addictive elements and work on them has a chance to increase the effectiveness of prevention.