Subject The growing popularity of e-commerce services is increasingly attracting the attention of their users to the services of recommendation engines. Collaborative filtering is one of the well-known recommendation methods that help customers select products of their interest. However, recommender systems are open to malicious attacks to promote or discredit certain products. For instance, the so-called shilling attacks cause a significant change in the rating of products in the social network, thereby causing significant material and moral damage. Objectives The purpose of the study is to increase the robustness of e-commerce systems and protect them against any manipulation by attackers and unscrupulous users; to develop computer algorithms to prevent the intrusion of fake recommendations into the recommendation system. Methods The study employs data mining techniques and machine learning methods. Results The paper considers characteristics of shilling attacks, describes their main models and parameters, and offers effective methods to detect fake profiles. I developed a computer algorithm of recommendations based on clustering as a powerful method to counter shilling attacks in social networks. I investigated its resistance to shilling attacks being the most popular among intruder, and analyzed the influence of various attack parameters on the results of its work. Conclusions Computer simulation of shilling attacks and analysis of their impact on the forecast of shift in recommendations showed that the algorithm is resistant to shilling attacks without significant impact of introduced malicious profiles on the work of the recommender system.
Agarwal Deepak K., Chen Bee-Chung. Statistical Methods for Recommender Systems. Cambridge University Press, 2016, 293 p.
Chung C.-Y., Hsu P.-Y., Huang S.H. P: A Novel Approach to Filter out Malicious Rating Profiles from Recommender Systems. Decision Support Systems, 2013, vol. 55, iss. 1, pp. 314–325. URL: Link
Zhou W., Wen J., Koh Y.S. et al. Shilling Attacks Detection in Recommender Systems Based on Target Item Analysis. PLoS ONE, 2015, no. 10(7). URL: Link
Jannach D., Zanker M., Felfernig A., Friedrich G. Recommender Systems: An Introduction. Cambridge University Press, 2010, 352 p.
Thorat P.B., Goudar R.M., Barve S. Survey on Collaborative Filtering, Content-based Filtering and Hybrid Recommendation System. International Journal of Computer Applications, 2015, vol. 110, no. 4, pp. 31–36. URL: Link
Azadjalal M.M., Moradi P., Abdollahpouri A., Jalili M. A Trust-Aware Recommendation Method Based on Pareto Dominance and Confidence Concepts. Knowledge-Based Systems, 2017, vol. 116, pp. 130–143. URL: Link
Deng S., Huang L., Xu G. Social Network-Based Service Recommendation with Trust Enhancement. Expert Systems with Applications, 2014, vol. 41, iss. 18, pp. 8075–8084. URL: Link
Kalai A., Zayani C.A. et al. Social Collaborative Service Recommendation Approach Based on User's Trust and Domain-Specific Expertise. Future Generation of Computer System, 2018, vol. 80, pp. 355–367. URL: Link
Li J., Chen C., Chen H., Tong C. Towards Context-Aware Social Recommendation via Individual Trust. Knowledge-Based Systems, 2017, vol. 127, pp. 58–66. URL: Link