Clustering and analyzing social network users behavior by combining personality traits and digital footprints

Digital social networks are becoming more and more popular, offering a massive platform for analyzing human behavior in computer-mediated contexts. Human behavior can be explored by analyzing the set of digital footprints left by people when interacting with social networks. Digital footprints can be classified into active and passive when produced unintentionally. This work seeks to identify user profiles in social networks from the grouping of behavior data in social networks, demographic data, and socio-affective information. Thus, the feasibility of creating meaningful groups is verified, as well as a qualitative and quantitative analysis of the groups produced is made available, in order to understand the quality of the groups formed and their validity in relation to the revised knowledge of personality psychology. More specifically, unsupervised learning algorithms (clustering) were employed. Although this work analyzes a small group of users (157 participants), correlations observed in the related bibliography can be verified, being the first step for future proposals in order to raise awareness about the relationship of social networks, personality computation, and its related fields.