Spam Detection for Youtube Comments

Abstract

Spamming is the use of messaging system, to send an unsolicited message. YouTube is one of the biggest sites for the user to get information. The best thing about the YouTube is the user can subscribe the channel, like or dislike the video and also giving opinion on the comment section on that video, and YouTube has attracted to increase the number of users. This attracts the spammers by spamming the comments. In this paper the spam comments on YouTube offers limited tools for comment moderation, so that the spam volume is shockingly increasing which leads to the owners to disable the comment section in their video. By using machine learning the comments are detected and prevented. Sklearn, numpy, pandas these are the some of the packages for understanding purpose of machine learning. YouTube spam comment has potential to spread malware through comment fields, which will exploit vulnerabilities in the user’s machines. There are many approaches to detect Spam such as Artificial intelligence (AI), Machine Learning (ML), cryptography and others. A you tube has an automatic spam filter; this is the challenge even for the classification method. YouTube classifier classifies the comments by using classification methods as very short, symbols, abbreviations, links.