• Prathamesh Paware
• Sahil Gargate
• Ashvini Kawade
• Shrushti Pawar

The profit promoted by Google in its spick-and-span video distribution platform YouTube has attracted a growing scope of users. However, such success has conjointly attracted malicious users that aim to self-promote their videos or bear viruses and malware. Since YouTube offers restricted tools for comment moderation, the spam volume is shockingly increasing that is leading house owners of known channels to disable the comments section in their videos. Automatic comment spam filtering on YouTube could be a challenge even for established classification ways since the messages square measure terribly short and infrequently rife with slangs, symbols, and elisions. During this work, we've evaluated many top-performance classification techniques for such purposes. The applied math analysis of results indicates that with 99.9% of confidence level Bernoulli Naive Bayes, Decision trees, Logistic Regression, Random forests, Linear and Gaussian SVM’s square measure statistically equivalent. Therefore, it's important to search out some way to notice these videos and report them before they're viewed by innocent user

Machine learning, Random Forests, Logistic Regression, Bernoulli Naïve Bayes, Decision trees, linear and Gaussian SVMs.

"YOUTUBE SPAM FILTER", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 3, page no.e328-e330, March-2022, Available :http://www.jetir.org/papers/JETIR2203441.pdf

"YOUTUBE SPAM FILTER", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 3, page no. ppe328-e330, March-2022, Available at : http://www.jetir.org/papers/JETIR2203441.pdf