FRPI: A novel Followee Recommendation approach based on Potential Interest in social networks

• Published in: 2021 International Conference on Information Science, Parallel and Distributed Systems (ISPDS) pp. 116 - 120
• Main Author: Xue, Zhengyuan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2021
• Subjects: Blogs; followee; followee recommendation; Information science; potential interest; Privacy; Reliability; Social networking (online)
• DOI: 10.1109/ISPDS54097.2021.00030

The future development of the network will largely depend on whether we can obtain high-quality and reliable information from the network. With the large-scale popularity of social networks in recent years, social networks have gradually become an indispensable part of people's daily communication. In real social networks such as Twitter, the problem of followee recommendations is a practical problem. At present, there are many methods to recommend followees for users, but none of them can perfectly solve the related problems. This paper proposes a novel followee recommendation approach called FRPI, based on users' potential interests in social networks. Firstly, a new calculation method of the potential interest of users in other users (i.e., U2UInterest) is given, and the calculation result of U2UInterest is used to solve the followee recommendation problem. Finally, the Top-k result is taken as the recommended followees for the target user. In addition, according to the relevant conclusions, this paper further discusses a new method to calculate the popularity of any user in social networks. The experiment uses the real data set Twitter for verification. The results show that the accuracy of the proposed approach FRPI based on potential interest is about 10 times higher than that of the random recommendation method, 14.6% higher than that of the Topology-based recommendation method, and 46.8% higher than that of the PageRank-based recommendation method.