Виявлення гібридних атак користувачів на рейтинги рекомендаційних систем

Abstract

In today's digital age, shilling attacks have become a growing concern in online communities. Shilling attacks involve the use of fake accounts or paid individuals to artificially promote a product or service, mislead consumers, and manipulate online discussions. As these attacks can undermine trust in online platforms and harm users, there is a pressing need to develop effective methods for detecting and preventing shilling attacks. This study aims to leverage machine learning techniques to detect and mitigate the impact of shilling attacks in online communities. The objective of my research is to develop and implement effective methods for detecting user attacks on ratings and recommendations within recommendation systems, using machine learning models.