To Discover Trolling Patterns in Social Media: Troll Filter

Abstract

Troll - the word itself defines everything about bullying or harassing someone. Trolling is an important problem in the online world. Emotions or feelings perform a vital role in successful and effective communication between humans. However, this human communication is getting worst if there is a group of people who enjoy targeting someone and trolling, this happens in different social media like Facebook and Twitter. In this paper tried to filter out comments, which are negative or insulting. Goal of this paper is to identify the targets of trolls, so as to prevent trolling before it happens. For this purpose used sentiment analysis (Positive or Negative) through machine learning. The major focus of this paper was on comparing different machine learning algorithms for the task of sentiment classification. For classification, many classifiers are available but results are very promising with Naive Bayes, Support Vector Machines (SVM) and Maximum Entropy (MaxEnt) classifiers. The major findings were evaluated that the Support Vector classifier provides the highest classification accuracy for this domain.