AutomatedDetectionOfHateSpeechTowardsWomanOnTwitter
Given the steadily growing body of social media content, hate speech towards women is increasing. Such kind of contents have the potential to cause harm and suffering on an individual basis, and they may lead to social tension and disorder beyond cyber space. To support the automatic detection of cyber hate online, specifically on Twitter, we build a supervised learning model which is developed to classify cyber hate towards woman on Twitter. Turkish tweets, with a hashtag specific to choice of clothing for women, have been collected and five machine learning based classification algorithms were applied including Support Vector Machines (using polynomial and RBF Kernel), J48, Naive Bayes, Random Forest and Random Tree. Preliminary results showed that hateful contents can be detected with high precision however more sophisticated approaches are necessary to improve recall. Keywords—Hate speech recognition, machine learning, classification, tf-idf
