一种基于节点相似性的链接预测算法

发布时间：2018-06-19 05:34

本文选题：社会网络 + 链接预测　；参考：《中国科技论文》2013年07期

【摘要】：针对社会网络中节点关系预测困难的问题,提出了一种新的链接预测算法:邻居关系权值算法。该算法将共同邻居节点与其他邻居节点之间的边赋予不同的权值,进而计算被预测节点之间的相似性。算法通过5个社会网络数据集进行实验,采用AUC(area under the receiver operating characteristic curve)指标和Precision指标评价其效果。实验表明新算法的预测准确率整体上高于已有的基于节点相似性的链接预测算法,同时该算法保持了较低的时间复杂度。
[Abstract]:A new link prediction algorithm is proposed to solve the problem of node relationship prediction in social networks . A new link prediction algorithm is proposed , which gives different weights to the edges between the common neighbor nodes and other neighbor nodes , and then calculates the similarity between the predicted nodes . The algorithm uses the area under the receiver operating characteristic curve index and the Precision index to evaluate its effect . The experiment shows that the prediction accuracy of the new algorithm is higher than that of the existing link prediction algorithm based on node similarity , and the algorithm keeps the lower time complexity .
【作者单位】：哈尔滨工程大学计算机科学与技术学院;
【基金】：高等学校博士学科点专项科研基金资助项目(20112304110011) 国家自然科学基金资助项目(61073041,61073043)
【分类号】：TP393.0

【参考文献】