摘要
链接预测是图数据挖掘中的一个重要问题。它是通过已知的网络结构等信息预测和估计尚未链接的两个节点存在链接的可能性。目前大部分基于节点相似性的链接预测算法只考虑共同邻居节点的个体特征,针对目前预测算法对共同邻居节点间相互关系的考虑不足,提出了一种新算法:节点引力指数算法。该算法在保持低时间复杂度的同时,提高了预测的准确率。通过多个现实网络实验证实了算法的预测效果。
Link prediction is an important issue in graph mining.It aimed at estimating the likelihood of the existence of links between nodes by the known network structure information.Currently,most link prediction algorithms based on node similarity consider only the individual characteristics of common neighbor nodes.We designed a new algorithm exploiting the interactions between common neighbors,namely Individual Attraction Index.While maintaining low time complexity,this algorithm remarkably improved the accuracy of prediction.This paper proved well the best overall performance of this new algorithm by comparing three well-known node similarity algorithms on eight real networks with Individual Attraction Index.
出处
《计算机科学》
CSCD
北大核心
2011年第7期162-164,199,共4页
Computer Science
基金
国家自然科学基金(60905025
61074128)
国家高技术研究发展计划(2009AA04Z136)资助
关键词
复杂网络
数据挖掘
链接预测
节点相似度
节点引力指数
Complex network
Data mining
Link prediction
Node similarity
Individual attraction index