ComRank: Joint Weight Technique for the Identification of Influential Communities

• 作 者: Muhammad Azam Zia Zhongbao Zhang Ximing Li Haseeb Ahmad Sen Su
• 机 构: State key Laboratory of Networking and Switching Technology,Beijing University of Posts and Telecomm
• 出 处: 中国通信
• 发表日期: 2017年04期
• 关键词: online social networks community rank citation network PageRank influence

Recently, the community analysis has seen enormous research advancements in the field of social networks. A large amount of the current studies put forward different models and algorithms about most influential people. However, there is little work to shed light on how to rank communities while considering their levels that are determined by the quality of their published contents. In this paper, we propose solution for measuring the influence of communities and ranking them by considering joint weight composed of internal and external influence of communities. To address this issue, we design a novel algorithm called Com Rank: a modification of Page Rank, which considers the joint weight in order to identify impact of each community and ranking them. We use real-world data trace in citation network and perform extensive experiments to evaluate our proposed algorithm. The comparative results depict significant improvements by our algorithm in community ranking due to the inclusion of proposed weighting feature.

Recently, the community analysis has seen enormous research advancements in the field of social networks. A large amount of the current studies put forward different models and algorithms about most influential people. However, there is little work to shed light on how to rank communities while considering their levels that are determined by the quality of their published contents. In this paper, we propose solution for measuring the influence of communities and rating them by considering joint weight composed of internal and external influence of communities. To address this issue, we design a novel algorithm called Com Rank: a modification of Page Rank, which considers the joint weight in order to identify impact of each community and rating them. We use real-world data trace in citation network and perform extensive experiments to evaluate our proposed algorithm. The comparative results depict significant improvements by our algorithm in community ranking due to the inclusion of p roposed weighting feature.