A Unified Deep Learning Architecture for Abuse Detection

Hate speech, offensive language, sexism, racism, and other types of abusive behavior have become a common phenomenon in many online social media platforms. In recent years, such diverse abusive behaviors have been manifesting with increased frequency and levels of intensity. Despite social media's efforts to combat online abusive behaviors this problem is still apparent. In fact, up to now, they have entered an arms race with the perpetrators, who constantly change tactics to evade the detection algorithms deployed by these platforms. Such algorithms, not disclosed to the public for obvious reasons, are typically custom-designed and tuned to detect only one specific type of abusive behavior, but usually miss other related behaviors. In the present paper, we study this complex problem by following a more holistic approach, which considers the various aspects of abusive behavior. We focus on Twitter, due to its popularity, and analyze user and textual properties from different angles of abusive posting behavior. We propose a deep learning architecture, which utilizes a wide variety of available metadata, and combines it with automatically-extracted hidden patterns within the text of the tweets, to detect multiple abusive behavioral norms which are highly inter-related. The proposed unified architecture is applied in a seamless and transparent fashion without the need for any change of the architecture but only training a model for each task (i.e., different types of abusive behavior). We test the proposed approach with multiple datasets addressing different abusive behaviors on Twitter. Our results demonstrate high performance across all datasets, with the AUC value to range from 92% to 98%.