Recognizing Hate-prone Characteristics of Online Hate Speech Targets

Abstract

Recognizing the characteristics of online hate speech targets can provide important information, which can help predict potential targets and protect them. Understanding the targets of online hate characteristics in the context of social media platforms where the hate is dramatically increasing lately can improve awareness of how different people react to online hate, which we anticipate will be manifested differently for different demographic groups. Targets of online hate play an important role in providing an extensive and informative explanation of the online hate event. This has been barely studied in previous research. In this PhD research, we propose a hate prone characteristics recognition framework for online hate targets, which consists of several modules, including data collection, data pre-processing, feature extraction, contextualisation and the hate prone characteristics recognition model that has the ability to recognise the common online hate prone characteristics to enhance the online hate prevention services, and finally, the hateful replies prediction model. This online hate prediction model has the potential to be personalised/adaptive in future applications.