Composite and Mutual Link Prediction using SVM in Social Networks
3055 Words13 Pages
A B S T R A C T
Link prediction is a key technique in many applications in social networks; where potential links between entities need to be predicted. Typical link prediction techniques deal with either uniform entities, i.e., company to company, applicant to applicant links, or non-mutual relationships, e.g., company to applicant links. However, there is a challenging problem of link prediction among the composite entities and mutual links; such as accurate prediction of matches on company dataset, jobs or workers on employment websites, where the links are mutually determined by both entities that composite entity belong to disjoint groups. The causes of interactions in these domains makes composite and mutual link prediction significantly different from the typical version of the problem. This work addresses these issues by proposing the Support Vector Machine model. By implementing the proposed algorithm it is expected that the accuracy will get increased in the link prediction problem.
Key words: Link prediction, Potential links, Composite, Mutual links, Support Vector Machine.
A social networking service is an online service, platform, or site that target to facilitating the building of social networks or social relations between people who, for example, share importance's, activities, backgrounds, or real-life communications. A social network service consists of a representation of each user, his/her social links, and a variety of new services. This is used to model the interaction among the communities on the social networks. Where the graphs are used to represent the interactions between those communities, in which nodes are representing to people in some communities and links are representing the association between those people.
Understanding the association between two specific nodes by predicting the likelihood of a future but not currently existing association between them is a fundamental problem known as link prediction.
Interaction on the social network involves both positive and negative relationships, e.g., since attempts to establish a relationship may fail due to decline from the expected target. This generates links that signify rejection of invitations, disapproval of applications, or expression of disagreement with others' opinions. Such social networks are mutual since the sign of a link indicating whether it is positive or negative depends on the attitudes or belief of both entities forming the link. Moreover, mutual positive and negative relationships have been even less investigated.
Fig.1: Collaborative Information for Composite and Mutual Link Prediction
Interaction Predicted Preference Predicted match
Recently, social network analysis has had a variety of applications, such as online dating sites, education admission portals as well as jobs, employment, career and hiring sites, where people in the networks have different roles and links between them can only between people in different roles.