Features Extraction for Link Prediction in Social Networks

The advent of social networks is one of the most exciting events in recent years. Its continuous development generated a huge information source which has been attracting attention of many researchers. Link prediction is an important problem in social network analysis. A link between two individuals is established base on many elements. Existing links give the individuals more chances to meet new people who can become their friends in future. In addition, personal features are necessary condition to form a stable link. Most present studies thoroughly mined network topology but used a little own information of users to recommend potential links. This affects to the quality of prediction. In this paper, we concentrate on the extraction of more features from user to improve the quality of predicted links. We proposed a measure calculating the similarity of individuals. The measure considers not only network topology but also the personal features. The higher similarity score, the more ability a new link appears. The experimentation shows that the simple combination of information as above brings considerable results.