PROJECT TITLE:

Friendbook A Semantic-Based Friend Recommendation System for Social Networks - 2015

ABSTRACT:

Existing social NetWorking services suggest friends to users based mostly on their social graphs, that could not be the foremost acceptable to reflect a user's preferences on friend choice in real life. In this paper, we gift Friendbook, a novel semantic-primarily based friend recommendation system for social networks, that recommends friends to users primarily based on their life designs rather than social graphs. By making the most of sensor-rich smartphones, Friendbook discovers life designs of users from user-centric sensor information, measures the similarity of life styles between users, and recommends friends to users if their life designs have high similarity. Inspired by Text Mining, we tend to model a user's daily life as life documents, from that his/her life styles are extracted by using the Latent Dirichlet Allocation algorithm. We additional propose a similarity metric to live the similarity of life styles between users, and calculate users' impact in terms of life designs with a devotee-matching graph. Upon receiving missive of invitation, Friendbook returns a listing of people with highest recommendation scores to the question user. Finally, Friendbook integrates a feedback mechanism to further improve the recommendation accuracy. We have a tendency to have implemented Friendbook on the Android-based mostly smartphones, and evaluated its performance on both little-scale experiments and giant-scale simulations. The results show that the recommendations accurately replicate the preferences of users in selecting friends.