Degree based Classification of Harmful Speech using Twitter Data - W18-4413

Harmful speech has various forms and it has been plaguing the social media in different ways.If we need to crackdown different degrees of hate speech and abusive behavior amongst it, theclassification needs to be based on complex ramifications which needs to be defined and holdaccountable for, other than racist, sexist or against some particular group and community. Thispaper primarily describes how we created an ontological classification of harmful speech basedon degree of hateful intent, and used it to annotate twitter data accordingly. The key contributionof this paper is the new dataset of tweets we created based on ontological classes and degrees ofharmful speech found in the text. We also propose supervised classification system for recogniz-ing these respective harmful speech classes in the texts hence. This serves as a preliminary workto lay down foundation on defining different classes of harmful speech and subsequent work willbe done in making it’s automatic detection more robust and efficient.