Anti social comment classification based on kNN algorithm

Abstract

Billions of contributions are made every day across multiple online communities and social media websites in the form of social messages, social blogs and online discussion. The aim of this paper is to identify such comments and posts which are racist and malicious in nature so that they could be effetely banned and removed in order to counter them. This article uses set of documents with racist comments as text corpus on which appropriate machine learning algorithm is applied to detect racist comments or meaning. To detect anti-social content there is a need to find the extent of similarity between a pair of text messages as a source and classified terms which are antisocial or in discriminating terms. The approach devised in this article to detect antisocial behavior is a technique based on term frequency based content classification.