PROJECT TITLE :

Cross-OSN User Modeling by Homogeneous Behavior Quantification and Local Social Regularization

ABSTRACT:

In the context of social media services, data shortage has severally hindered correct user modeling and sensible personalised applications. This paper is motivated to leverage the user knowledge distributed in disparate on-line social networks (OSN) to make up for the info shortage in user modeling, that we tend to confer with as “cross-OSN user modeling.” Typically, the data that the identical user distributes in numerous OSNs consist of both behavior knowledge (i.e., interaction with multimedia things) and social data (i.e., interaction between users). This paper focuses on the following two challenges: one) the way to combination the users’ cross-OSN interactions with multimedia items of the same modality, that we tend to decision cross-OSN homogeneous behaviors, and a pair of) a way to integrate users’ cross-OSN social information with behavior data. Our proposed answer to address the challenges consist of two corresponding parts as follows. one) Homogeneous behavior quantification, where homogeneous user behaviors are quantified based mostly on their importance in reflecting user preferences. After quantification, the examined cross-OSN user behaviors are aggregated to construct a unified user-item interaction matrix. 2) Local social regularization, where the cross-OSN social information is integrated as regularization in matrix factorization-based user modeling at native topic level. The proposed cross-OSN user modeling resolution is evaluated in the appliance of customized video recommendation. Carefully designed experiments on self-collected Google+ and YouTube datasets have validated its effectiveness and the advantage over single-OSN-primarily based methods.