结合网络结构和非线性恢复提高信誉评估性能
Improving the performance of reputation evaluating by combining the structure of network and nonlinear recovery
摘要：
评价者的声誉特征对于消费者从在线评价系统中获取有用信息具有重要意义。此外，如何克服垃圾邮件攻击对评分系统的影响，并获得可信的评价者信誉也是本研究的一个重要课题。我们注意到，现有的评价者声誉评价方法大多只依赖于评价者的评分信息和异常行为来建立声誉系统，而忽略了评分系统的系统性方面，包括评价者-客体二分网络的结构和非线性效应的影响。本研究将评价者-客体二分网络的结构与评价信息相结合，并引入惩罚因子和奖励因子，提出一种改进的信誉评价方法。在一个大规模人工数据集和两个真实数据集上对该方法进行了实证分析。实验结果表明，该方法在垃圾邮件攻击下具有更高的准确性和鲁棒性。这一新的思想为稀疏二分评价网络中信誉评价模型的建立提供了一种新的思路。
英文摘要：
Characterizing the reputation of an evaluator is particularly significant for consumer to obtain useful information from online rating systems. Furthermore, to overcome the difficulties with spam attacks on the rating system and to get the reliable on reputation of evaluators is an important topic in the research. We have noticed that most of the existing evaluator reputation evaluation methods only rely on the evaluator's rating information and abnormal behavior to establish a reputation system, which miss the systematic aspects of the rating systems including the structure of the evaluator-object bipartite network and the effects of nonlinear effects. This study we propose an improved reputation evaluation method by combining the structure of the evaluator-object bipartite network with rating information and introducing penalty and reward factors. This novel method has been empirically analyzed on a large-scale artificial data set and two real data sets. The results show that the proposed method is more accurate and robust in the presence of spam attacks. This fresh idea contributes a new way for building reputation evaluation models in sparse bipartite rating network.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
