A Unified Deep Learning Architecture for Abuse Detection

By Antigoni-Maria Founta, Despoina Chatzakou, Nicolas Kourtellis, Jeremy Blackburn, Athena Vakali, Ilias Leontiadis. 2018

250 citations recorded by the source corpus.

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%.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Fighting Offensive Language on Social Media with Unsupervised Text Style Transfer (2018, background): “Most previous work that address the problem of offensive language on social media has focused on text classification using different machine learning methods (Xiang et al., 2012; Warner and Hirschberg, 2012; Kwok and Wang, 2013; Wang et al., 2014; Burnap and Williams, 2015; Nobata et al., 2016; Davi”
Joint Modelling of Emotion and Abusive Language Detection (2020, background): “Founta et al. (2019) used transfer learning to fine-tune features from the author metadata network to improve abuse detection.”
A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Media (2019, background): “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.”
The Linguistic Ideologies of Deep Abusive Language Classification (2018, result): “While some earlier research in the classiﬁcation of abusive language used feature-based classiﬁcation techniques such as support vector machines (Warner and Hirschberg, 2012), we were interested in evaluating deep learning methods comparable to work such as Founta et al. (2018).”
Predictive Embeddings for Hate Speech Detection on Twitter (2018, methodology): “Founta et al. (2018) built a classiﬁer composed of two separate neural networks, one for the text and the other for metadata of the Twitter user, that were trained jointly in interleaved fashion.”
Aggression Detection in Social Media: Using Deep Neural Networks, Data Augmentation, and Pseudo Labeling (2018, background): “Rather than rely only on the textual content of social media posts in identifying hate speech, (Qian et al., 2018; Founta et al., 2018) investigated the use of metadata about the users or the posts.”
DeepHate: Hate Speech Detection via Multi-Faceted Text Representations (2020, methodology): “Founta [15] trains a combined RNN and Multilayer perceptron (MLP) on textual content and meta-information to perform hate speech detection.”
Towards Ethics by Design in Online Abusive Content Detection (2020, background): “However, the accuracy of lexicon and feature-based systems is often signiﬁcantly lower than the accuracy of deep learning models (Dixon et al., 2018; Gunasekara and Nejadgholi, 2018; Founta et al., 2019).”