A Novel Approach for Youtube Video Spam Detection using Markov Decision Process
Simran Kanodia, Rachna Sasheendran, Vinod Pathari · 2018
Social networking websites have become an integral part of the day to day lives of people. People turn to social media for interacting with other people, sharing ideas, gaining knowledge, for entertainment and staying informed about the events happening in the rest of the world. Among these sites, YouTube has emerged as the most popular website for sharing and viewing video content. This popularity of YouTube has also attracted spammers, who upload videos with the sole purpose of polluting the system content and causing dissatisfaction among other viewers. These spam videos may be unrelated to their title or may contain pornographic content. Therefore, it is very important to find a way to detect these videos and report them before they are viewed by innocent users. In this paper, we propose a Markov Decision Process approach to model the problem of YouTube video spam detection. We analyze the accuracy of the policy returned by the model and compare it with the accuracy of other data mining algorithms that have been proposed for video spam detection. We find that the proposed model gives a superior performance than the other models.