Proposed Techniques to Remove Flaming Problems from Social Networking Sites and outcome of Naïve Bayes Classifier for Detection of Flames

Natural Language Processing (NLP)[1][2][5] is a field of Computer Science concerned with the interactions between Computer and Human (Natural) Languages. Social Networking Sites are amongst the most effective communication tools now a days. But it also gave rise to the problem of flaming which is difficult to deal with. A flaming incident is triggered by comments and actions of users in SNS that ends up damaging reputation or causing negative impact on the target party. So in this paper, in order to prevent such damage, Semi-Supervised Learning[1] Approach is presented. It includes a proposed architecture that involves Naïve Bayes Classifier, Maximum Entropy Classifier and Feature Selection based on Entropy method. This paper also includes application and result of Naïve Bayes Classifier, trained so as to detect the class of the input comments as whether it belong to positive, negative or neutral class. Sentiment Analysis[1] Dataset containing tweets and there polarity class(positive, negative or neutral) has been used. A training set of 17282 tweets has been used to train Naïve Bayes Classifier which was able to correctly classify 70% of the test set of 4332 tweets.