排名聚合算法在少量长列表聚合中的性能比较分析

排名聚合将多个排名列表聚合成一个综合排名列表,可应用于推荐系统、链路预测、元搜索、提案评选等.当前已有工作从不同角度对不同排名聚合算法进行了综述、比较,但存在算法种类较少、数据统计特性不清晰、评价指标不够合理等局限性.不同排名聚合算法在提出时均声称优于已有算法,但是用于比较的方法不同,测试的数据不同,应用的场景不同,因此何种算法最能适应某一任务在很多情况下仍不甚清楚.本文基于Mallows模型,提出一套生成统计特性可控的不同类型的排名列表的算法,使用一个可应用于不同类型排名列表的通用评价指标,介绍9种排名聚合算法以及它们在聚合少量长列表时的表现.结果发现启发式方法虽然简单,但是在排名列表相似度较高、列表相对简单的情况下,能够接近甚至超过一些优化类方法的结果;列表中平局数量的增长会降低聚合排名的一致性并增加波动;列表数量的增加对聚合效果的影响呈现非单调性.整体而言,基于距离优化的分支定界方法 (FAST)优于其他各类算法,在不同类型的排名列表中表现非常稳定,能够很好地完成少量长列表的排名聚合.     

A novel similarity measure for mining missing links in long-path networks

Network information mining is the study of the network topology, which may answer a large number of application-based questions towards the structural evolution and the function of a real system. The question can be related to how the real system evolves or how individuals interact with each other in social networks. Although the evolution of the real system may seem to be found regularly, capturing patterns on the whole process of evolution is not trivial. Link prediction is one of the most important technologies in network information mining, which can help us understand the evolution mechanism of real-life network. Link prediction aims to uncover missing links or quantify the likelihood of the emergence of nonexistent links from known network structures. Currently, widely existing methods of link prediction almost focus on short-path networks that usually have a myriad of close triangular structures. However, these algorithms on highly sparse or long-path networks have poor performance. Here, we propose a new index that is associated with the principles of structural equivalence and shortest path length (SESPL) to estimate the likelihood of link existence in long-path networks. Through a test of 548 real networks, we find that SESPL is more effective and efficient than other similarity-based predictors in long-path networks. Meanwhile, we also exploit the performance of SESPL predictor and of embedding-based approaches via machine learning techniques. The results show that the performance of SESPL can achieve a gain of 44.09% over GraphWave and 7.93% over Node2vec. Finally, according to the matrix of maximal information coefficient (MIC) between all the similarity-based predictors, SESPL is a new independent feature in the space of traditional similarity features.