Detecting shilling attacks in recommender systems based on analysis of user rating behavior

Gray-Box Shilling Attack: An Adversarial Learning Approach

Recommender systems are essential components of many information services, which aim to find relevant items that match user preferences. Several studies have shown shilling attacks can significantly weaken the robustness of recommender systems by injecting fake user profiles. Traditional shilling attacks focus on creating hand-engineered fake user profiles, but these profiles can be detected effortlessly by advanced detection methods. Adversarial learning, emerged in recent years, can be leveraged to generate powerful and intelligent attack models. To this end, in this paper, we explore potential risks of recommender systems and shed light on a gray-box shilling attack model based on generative adversarial networks, named GSA-GANs. Specifically, we aim to generate fake user profiles that can achieve two goals: unnoticeable and offensive. Towards these goals, there are several challenges that we need to address: (1) learn complex user behaviors from user-item rating data; (2) adversely influence the recommendation results without knowing the underlying recommendation algorithms. To tackle these challenges, two essential GAN modules are respectively designed to make generated fake profiles more similar to real ones and harmful to recommendation results. Experimental results on three public datasets demonstrate that the proposed GSA-GANs framework outperforms baseline models in attack effectiveness, transferability, and camouflage. In the end, we also provide several possible defensive strategies against GSA-GANs. The exploration and analysis in our work will contribute to the defense research of recommender systems.

Artificial intelligence in E-Commerce: a bibliometric study and literature review

This paper synthesises research on artificial intelligence (AI) in e-commerce and proposes guidelines on how information systems (IS) research could contribute to this research stream. To this end, the innovative approach of combining bibliometric analysis with an extensive literature review was used. Bibliometric data from 4335 documents were analysed, and 229 articles published in leading IS journals were reviewed. The bibliometric analysis revealed that research on AI in e-commerce focuses primarily on recommender systems. Sentiment analysis, trust, personalisation, and optimisation were identified as the core research themes. It also places China-based institutions as leaders in this researcher area. Also, most research papers on AI in e-commerce were published in computer science, AI, business, and management outlets. The literature review reveals the main research topics, styles and themes that have been of interest to IS scholars. Proposals for future research are made based on these findings. This paper presents the first study that attempts to synthesise research on AI in e-commerce. For researchers, it contributes ideas to the way forward in this research area. To practitioners, it provides an organised source of information on how AI can support their e-commerce endeavours.

Detecting Shilling Attacks Using Hybrid Deep Learning Models

Recommendation systems play a significant role in alleviating information overload in the digital world. They provide suggestions to users based on past symmetric activities or behaviors. Being heavily dependent on users’ behavior, they tend to be vulnerable to shilling attacks. Therefore, protecting them from attacks’ effects is highly important. As shilling attacks have features of a large number of ratings and increasing complexity in attack models, deep learning methods become proper alternatives for more accurate attack detections. This paper proposes a hybrid model of two different neural networks, convolutional and recurrent neural networks, to detect shilling attacks efficiently. The proposed deep learning model utilizes the transformed network architecture for undertaking the attributes derived from user-rated profiles. This architecture enables modeling of the temporal and spatial information in the recommendation system’s ratings. The hybrid model overcomes the limitations of the existing shilling attack deep-learning methods to enhance the recommendation systems’ efficiency and robustness. Experimental results show that the hybrid model results in better predictions on the Movie-Lens 100 K and Netflix datasets by accurately detecting most of the obfuscated attacks compared to the state-of-art deep learning algorithms used for investigation.