A demographic-based approach for improved content categorization in social networking

Social networks enable users to freely communicate with each other and share their personal and social data, ongoing activities, interests, preferences, and views about different topics. Users demographics information plays an essential role in their interests and preferences identification. For example, women have not the same interests as men; women interest more about fashion and makeup than men, where, men interest more about sports than women. Therefore, the demographic attributes such as gender, age, location, marital status, education, and career can affect user's interests and preferences. In this paper, we propose a new approach which incorporates users' demographic attributes on the content classification. The experiments are done on a large Facebook dataset in order to analyze the effect of these demographic attributes on the performance of the categorization of the shared textual content in social networks.