An intelligent Approach for Educational Levels Classification in Social Networks

Abstract

The identification of social groups remains one of the main analytical themes in the analysis of social networks and, in more general terms, in the study of social organizations. Research at the intersection of machine learning and the social sciences provides critical new insights into social organizations. In building of machine learning model for social data, our goal is to predict, describe, and/or explain some social phenomenon. In this work we propose a relatively novel approach to the classification and identification of educational levels of the social network users through the texts. This work can be used for various social, political, cultural, religious, and economic purposes, as well as for security purposes to protect society from crime, violence and social ills. In this approach, the classification will depend on proposing new features as well as appropriate dictionaries for the language or the dialect considered (Arabic language - Algerian dialect) we used four of the learning algorithms (SVM,DT,RF,NB). Accurate results and analytical framework will be provided in this paper .