Research On Friend Recommendation On Microblog Based On Mapreduce

The development of Internet technology promotes the social progress. With the popularizing of computer and network, people’s daily life turn to be more digital and electronic. The massive emerging data makes the limitations of the single chip processing more obvious. In that case, computing power and data storage capacity of the machine are urged to be improved. However, in terms of massive data the performance of the single chip was still very limited, so the cloud computing emerges and plays a very important role. Social networks as a new projections of social media, especially, the emergence of microblog has become an important part of human’s daily life. The behavior model of microblog users, who are considered as vital component of microblog, is the hotspot in this field. In this paper, we focused on the evaluation of users’ influence and friend recommendation in microblog social network, and proposed a modified algorithm and model with the help of MapReduce to complete the distributed transformation.In this paper, an evaluation algorithm based on HRank was proposed to evaluate the users’ influence in microblog social networking platform. Basing on the user’s followers and their microblog forwarding numbers, two new H-index models of followers H-index and microblog-forwarded H-Index were given. And then the HRank model was established to make comprehensive assessment on users’influence. Finally, the experiments were done to analyze the correlation on users’influence rank between the results given by the HRank model, the PageRank model, and the model given by Sina microblog. Results show that the HRank model gains higher correlation, the value of which improves about 10% than PageRank model. It indicates that the HRank model could be used to identify users influence effectively.In view of the friend recommendation problem in social networks, a friend recommendation algorithm based on the theory of the three-degree influence is proposed in this paper. The relationships between social-network-users are not only including the mutual friends, but also the other connecting relations with different length. By introducing the theory of three-degree influence, the algorithm takes all the relationships within three-degree between users into account, while not only considering the number of mutual friends between users as the main basis of the friend recommendation. The experiments on Sina microblog and Facebook show that the precision and recall rate are improved by about 5% and 0.8% respectively than merely based on mutual friends. It indicates the better recommendation performance for the improved recommendation algorithm. It could be helpful for the social platform to improve the recommendation system and enhance the users’experience.The friend recommendation algorithm based on the theory of the three-degree influence only considered the network structural property of users’ friends relationship. And users’ behavior property was ignored, which was reflected by the users’ influence. In this paper, we bring users’ influence in the process of friend recommendation based on the theory of the three-degree influence, and discuss the algrithm’s effectiveness after considering the user’s influence.