A Unified Deep Learning Architecture for Abuse Detection

Abstract

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. This is due to the openness and willingness of popular media platforms, such as Twitter and Facebook, to host content of sensitive or controversial topics. However, these platforms have not adequately addressed the problem of online abusive behavior, and their responsiveness to the effective detection and blocking of such inappropriate behavior remains limited. 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 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. To make the approach tangible, we focus on Twitter data 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. We apply this unified architecture in a seamless, transparent fashion to detect different types of abusive behavior (hate speech, sexism vs. racism, bullying, sarcasm, etc.) without the need for any tuning of the model architecture for each task. We test the proposed approach with multiple datasets addressing different and multiple abusive behaviors on Twitter. Our results demonstrate that it largely outperforms the state-of-art methods (between 21 and 45% improvement in AUC, depending on the dataset).1 Did trolls cost Twitter 3.5bn and its sale? goo.gl/PlIL66 2 A Calendar of Our Safety Work (Twitter): https:

A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Media

“…Waseem et al [23] brought a new insight to hate speech and abusive language detection tasks by proposing a multi-task learning framework to deal with datasets across different annotation schemes, labels, or geographic and cultural influences from data sampling. Founta et al [7] built a unified classification model that can efficiently handle different types of abusive language such as cyberbullying, hate, sarcasm, etc. using raw text and domain-specific metadata from Twitter.…”

Towards Integration of Statistical Hypothesis Tests into Deep Neural Networks

“…Our model is also similar to the one by Founta et al (2018) with two differences. First, they assume that classes are provided with metadata, while our model extracts class descriptions directly from the training data.…”

“…The consequence of this decision is that the probability for a class is not independent of the other class probabilities, which would not be the desired behavior when dealing with a multi-label classification task. For instance, in a multi-label classification for hate speech detection, the probability of a Figure 1: The system architecture comment for being offensive is independent of its probability of being hateful, because an offensive tone can be used in a text that is not necessarily hateful (Founta et al, 2018). For this reason, instead of Softmax, we use the Sigmoid activation function…”

“…Text classification is a complex problem in Natural Language Processing (NLP) with lots of applications from sentiment analysis (Liu, 2015) to question answering (Aghaebrahimian and Jurčíček, 2016b,a;Yu et al, 2014) or abusive language detection (von Grünigen et al, 2018;Founta et al, 2018), to name just a few. Text classification is defined as the task of assigning a certain pre-defined class to a document.…”

NLPR@SRPOL at SemEval-2019 Task 6 and Task 5: Linguistically enhanced deep learning offensive sentence classifier

“…We checked the graphemic characteristics of the written text and we gave this as a feature to the model. We mainly used the non user related features defined in Founta et al (2018).…”