A hybrid recommender system based non-common items in social media

A decade ago, an increasing number of people are accustomed to utilize the social media for sharing their impressions, experiences, information, and resources. A social media website does not only provide the information sharing functions for the users but also has the capability to match users' similar interesting preferences. Therefore, a recommender that can recommend highly interesting information to the users will make the users easy to gain new information they want, suggest implicit friends who have same hobbies, and increase the interactions with others. In this paper, we proposed a weight based recommendation method for social media, which usually has only the personal items uploaded by the users. Our evaluation results show this recommendation approach is feasible and flexible. Furthermore, we applied the proposed recommendation method into our information sharing system and look forward it can raise the interactions with the users in order to improve the quality of community.