【摘要】 由于组织与组织之间,个人与个人之间的社会关系在社会生活中不断变化,因此在不同领域之间形成了动态的社会网络结构。在理解网络这种动态的性质以及确定未来的关系方面,链路预测是一个重要且有效的解决方法。它运用网络当前状态的结构特征去预测将来网络节点之间可能存在的关联。论文基于药物推荐的应用特征,提出一种面向疾病-药物网络的链路预测方法。当前大多数链路预测的研究都是基于单一模式的网络结构。区别与单一模式网络,论文提出基于二分网络(如疾病-药物网络)的链路预测算法。同时为了验证提出算法的预测效果,论文选择四个经典的链路预测算法进行对比。实验结果显示,论文提出的方法比其他的基于链路预测的类似方法成功率更高。更多还原

【Abstract】 The social relations between organizations are constantly changing,which leads to the formation of dynamic social network structure between different fields. Link prediction is an effective and important way to identify the future relations and to understand the dynamic nature of networks. Link prediction uses the structural characteristics of the network to predict future links between nodes. Based on the application characteristics of the drug recommendation,a link prediction method for the disease-drug network is proposed in this paper. At present,most of the link prediction algorithms are based on the single mode network structure.Different from the single mode network,link prediction algorithm based on the bipartite network,such as disease-drug network,is proposed in this paper. At the same time,in order to verify the prediction effect of the proposed algorithm,four classical link prediction algorithms are chosen to compare. The experimental results show that the proposed method has a better predictive effect than the other link prediction methods based on structural similarity.更多还原