An Obfuscated Attack Detection Approach for Collaborative Recommender Systems

摘要

In the recent times, we have loads and loads of information available over the Internet. It has become very cumbersome to extract relevant information out of this huge amount of available information. So to avoid this problem, recommender systems came into play, which can predict outcomes according to user's interests. Although recommender systems are very effective and useful for users, the most used type of recommender system, i.e. collaborative filtering recommender system, suffers from shilling/profile injection attacks in which fake profiles are inserted into the database in order to bias its output. With this problem in mind, we propose an approach to detect attacks on recommender systems using Random Forest Classifier and find that, when tested at 10% attack, our approach outperformed earlier proposed approaches.