计算机工程与应用Issue(10)：150-154,159,6.DOI:10.3778/j.issn.1002-8331.1512-0268
基于非负矩阵分解的托攻击检测算法
Shilling attacks detection algorithm based on nonnegative matrix factorization
摘要
Abstract
The existing unsupervised detection algorithms have a high misjudgment rate for normal user. To solve this problem, a method for detecting shilling attack based on nonnegative matrix factorization is proposed. Firstly, features of the user are extracted from the nonnegative matrix factorization technique. Secondly, the K-means clustering method is used to extract the initial normal user set and the initial shilling attack set. Finally, using features of initial normal user set classifies the initial shilling attack set, to detect shilling attacks. Experimental results show that this algorithms are more effective in detecting the attacks compared to other algorithms.关键词
推荐系统/非负矩阵分解/托攻击/检测算法Key words
recommender systems/nonnegative matrix factorization/shilling attack/detection algorithm分类
信息技术与安全科学引用本文复制引用
方楷强,王靖..基于非负矩阵分解的托攻击检测算法[J].计算机工程与应用,2017,(10):150-154,159,6.基金项目
国家自然科学基金(No.61370006) （No.61370006）
福建省自然科学基金(No.2014J01237) （No.2014J01237）
福建省教育厅科技项目(No.JA12006) （No.JA12006）
福建省高等学校新世纪优秀人才支持计划(No.2012FJ-NCET-ZR01) （No.2012FJ-NCET-ZR01）
华侨大学中青年教师科技创新资助计划(No.ZQN-PY116). （No.ZQN-PY116）
