A survey of attack detection approaches in collaborative filtering recommender systems
- PDF / 5,851,860 Bytes
- 56 Pages / 439.37 x 666.142 pts Page_size
- 19 Downloads / 326 Views
A survey of attack detection approaches in collaborative filtering recommender systems Fatemeh Rezaimehr1 · Chitra Dadkhah1
© Springer Nature B.V. 2020
Abstract Nowadays, due to the increasing amount of data, the use of recommender systems has increased. Therefore, the quality of the recommendations for the users of these systems is very important. One of the recommender systems models is collaborative filtering (CF) which uses the ratings given by the users to the items. But many of these ratings may be noisy or inaccurate so they reduce the quality of the recommendations. Sometimes users, using fake profiles, try to change the recommendations in their favor. Since satisfaction and trust in such systems are very important and useful, it would be better to find a way to identify these types of users. Despite numerous studies on CF recommender systems, the design of a robust recommender system is still a challenging problem. In this paper, we have analyzed the 25 previous samples of research on collaborative filtering recommender system (CFRS) for attack detection from 2009 to 2019. Most of these papers focus mainly on movie recommendations. According to these analyzes, we have categorized attack detection methods on CFRS in four categories: clustering, classifying, feature extraction and probabilistic approaches. The evaluation measures, the dataset, and attacks features used in the attack detection approaches are discussed. Keywords Collaborative filtering · Recommender systems · Fake user · Detecting attack · Clustering · Feature extraction · Shilling attack
1 Introduction Recommender systems help the users find the relevance items among the vast information on the web based on their preferences in a short time using their activities or items properties. Aggarwal (2016) categorized the recommender systems approaches into four basic models: collaborative filtering (CF), content-based (CB), knowledge-Based (KB) and ensemble-based and hybrid (HB). CF model only analyze historical users transactions for recommendation, while the recommendation of CB model are based on items * Chitra Dadkhah [email protected] Fatemeh Rezaimehr [email protected] 1
Computer Engineering Faculty, K.N.Toosi University of Technology, Tehran, Iran
13
Vol.:(0123456789)
F. Rezaimehr, C. Dadkhah
attributes. In KB model, the recommendations are based on explicit user requirements. In HB model, the different models of recommender systems are combined to make recommendations more accurate. In CFRS, to provide recommendations for the target user, similar users are find, and then based on their activities recommendation are made. The target/active user is a user who will be given a recommendation. CF model are classified into two approaches: memory-based and model-based. In memory-based approaches, the user–item interaction matrix (e.g., historical ratings) is used to find the similarity between users/items (Sarwar et al. 2001; Moradi et al. 2016; Rezaeimehr et al. 2018). On the other hand, in model-based approaches, such
Data Loading...
