Re-scale AdaBoost for attack detection in collaborative filtering recommender systems

摘要

Collaborative filtering recommender systems (CFRSs) are the key components of successful E-commerce systems. However, CFRSs are highly vulnerable to “shilling” attacks or “profile injection” attacks due to its openness. Since the size of attackers is usually far smaller than genuine users, conventional supervised learning based detection methods could be too “dull” to handle such imbalanced classification. In this paper, we improve detection performance from following two aspects. Firstly, we extract well-designed features from user profiles based on the statistical properties of the diverse attack models, making hard detection scenarios become easier to perform. Then, refer to the general idea of re-scale Boosting (RBoosting) and AdaBoost, we apply a variant of AdaBoost, called the re-scale AdaBoost (RAdaBoost) as our detection method based on the extracted features. Finally, a series of experiments on the MovieLens-100K dataset are conducted to demonstrate the outperformance of RAdaBoost over other competing techniques such as SVM, kNN and AdaBoost.