Open AccessSpammer Detection: A Study of Spam Filter Commentson YouTube Videos
Author(s) -
Rafaqat Alam Khan
Publication year - 2018
Publication title -
international journal for electronic crime investigation
Language(s) - English
Resource type - Journals
eISSN - 2616-6003
pISSN - 2522-3429
DOI - 10.54692/ijeci.2019.030125
Subject(s) - spamming , computer science , monetization , spambot , publicity , filter (signal processing) , forum spam , world wide web , internet privacy , information retrieval , computer security , the internet , business , marketing , economics , computer vision , macroeconomics
This paper presents a methodology to find out the spam comments on YouTube videos. The purpose of this research is to find out the comments of those spam users, who comment for their own promotional intentions or to detect users whose comments that have no relevancy with the given video.The monetization policy introduced by YouTube for its user's channel and advertisement of different ads on YouTube videos has attracted a large number of users. This increase in a large number of users has also lead to an increase in malicious users whose job is to create automated bots for commenting and subscription to different YouTube channels. These malicious users'comments hurt the channel publicity and also affect the normal user's experience. YouTube is also working on this issue by using different methods to limit these kinds of automated bots malicious comments by blocking those comments. These kinds of methods are ineffective so far as spammers have found out different methods to bypass those heuristic approaches. Different machine learning approaches provide somehow better classification accuracy with the introduction of new approaches to solve it better than that. In this work, different techniques used for classification of spam comments with those of normal user comments to improve the classification and recent trend going on in this area are briefly analyzed to tackle this major issue.