Predicting the growth of new links by new preferential attachment similarity indices

Pramana

Ju Xiang²  Hui-Jia Li³  Yi Tang¹  Wan-Chun Yang⁵  Ke Hu¹  Xiao-Ke Xu3⁴ 

[1] Hunan Key Laboratory for Micro-Nano Energy Materials and Devices, and Laboratory for Quantum Engineering and Micro-Nano Energy Technology, Xiangtan University, Xiangtan 411105, China$$;Department of Basic Sciences, The First Aeronautical Institute of the Air Force, Xinyang 464000, China$$;School of Management Science and Engineering, Central University of Finance and Economics, Beijing 100080, China$$;College of Information and Communication Engineering, Dalian Nationalities University, Dalian 116605, China$$;College of Information Engineering, Xiangtan University, Xiangtan 411105, China$$

关键词: Link prediction;    preferential attachment;    network evolution.;   

DOI  :  

学科分类：物理（综合）

来源: Indian Academy of Sciences

【 摘 要 】

By revisiting the preferential attachment (PA) mechanism for generating a classical scale-free network, we propose a class of novel preferential attachment similarity indices for predicting future links in evolving networks. Extensive experiments on 14 real-life networks show that these new indices can provide more accurate prediction than the traditional one. Due to the improved prediction accuracy and low computational complexity, these proposed preferential attachment indices can be helpful for providing both instructions for mining unknown links and new insights to understand the underlying mechanisms that drive the network evolution.

【 授权许可】

Unknown