摘要
针对现有的用户概貌攻击检测算法在检测模糊攻击时精确度不高的问题,本文提出一种基于局部密度的用户概貌攻击检测算法.首先,利用LOF离群点检测算法为每个用户计算局部离群因子,得到用户的局部离群程度;然后,结合攻击用户对目标项目的评分与真实用户评分之间的差异,进一步确定目标项目及攻击目的,最终给出目标项目所对应的攻击概貌.实验结果表明,该算法无论是针对标准攻击还是模糊攻击,均具有较高的检测精度.
The existing user profile attacks detection algorithms have lower precision when detecting obfuscated attacks. With this in mind, a local density-based algorithm to detect user profile attacks is proposed. We first calculate the local outlier factor for each user profile using LOF outlier detection algorithm and get the local deviation degree of the users. Then combined with the difference of the target item rated by attackers and genuine users, we can find the target item and attack purpose so as to identify the corresponding attack profiles. The experimental results show that the proposed algorithm has higher accuracy both in detecting standard attack and obfuscated attack.
出处
《小型微型计算机系统》
CSCD
北大核心
2013年第4期850-855,共6页
Journal of Chinese Computer Systems
基金
河北省自然科学基金项目(F2011203219)资助
河北省高等学校科学技术研究重点项目(ZH2012028)资助
关键词
用户概貌攻击
攻击检测
模糊攻击
局部密度
局部离群因子
user profile attacks
attack detection
obfuscated attack
local density
local outlier factor
