微计算机信息
微計算機信息
미계산궤신식
CONTROL & AUTOMATION
2010年
6期
1-2
,共2页
协同过滤%推荐系统%描述文件注入攻击%攻击检测
協同過濾%推薦繫統%描述文件註入攻擊%攻擊檢測
협동과려%추천계통%묘술문건주입공격%공격검측
Collaborative Filtering%Recommender Systems%Attack Detection%Profile Injected Attacks
协同推荐系统广泛地应用于电子商务和信息访问系统,为新用户提供个性化的产品建议.然而,协同推荐系统存在着严重的安全隐患,使得恶意用户能够注入伪造的描述文件,影响或破坏提供给其他用户的推荐建议.本文探讨了检测响应描述文件注入攻击的方法,改进了协同过滤推荐算法,设计了基于检测响应方法的安全协同推荐系统框架.
協同推薦繫統廣汎地應用于電子商務和信息訪問繫統,為新用戶提供箇性化的產品建議.然而,協同推薦繫統存在著嚴重的安全隱患,使得噁意用戶能夠註入偽造的描述文件,影響或破壞提供給其他用戶的推薦建議.本文探討瞭檢測響應描述文件註入攻擊的方法,改進瞭協同過濾推薦算法,設計瞭基于檢測響應方法的安全協同推薦繫統框架.
협동추천계통엄범지응용우전자상무화신식방문계통,위신용호제공개성화적산품건의.연이,협동추천계통존재착엄중적안전은환,사득악의용호능구주입위조적묘술문건,영향혹파배제공급기타용호적추천건의.본문탐토료검측향응묘술문건주입공격적방법,개진료협동과려추천산법,설계료기우검측향응방법적안전협동추천계통광가.
A recommender system makes personalized product suggestions by extracting knowledge from the previous user interactions. However, recommenders raise serious security issues. Malicious users can bias or sabotage the recommendations that are provided to other users. Since collaborative recommender systems must be open to user input, it is difficult to design a system that cannot be at-tacked. In this paper, detection and response methods were discussed, then improved on collaborative filtering algorithm, designed a framework of secure collaborative recommendation system based on detection and response.
