The detection of abusive or offensive remarks in social texts has received significant attention in research. In several related shared tasks, BERT has been shown to be the state-of-the-art. In this paper, we propose to utilize lexical features derived from a hate lexicon towards improving the performance of BERT in such tasks. We explore different ways to utilize the lexical features in the form of lexicon-based encodings at the sentence level or embeddings at the word level. We provide an extensive dataset evaluation that addresses in-domain as well as cross-domain detection of abusive content to render a complete picture. Our results indicate that our proposed models combining BERT with lexical features help improve over a baseline BERT model in many of our in-domain and cross-domain experiments.

Swearing plays an ubiquitous role in everyday conversations among humans, both in oral and textual communication, and occurs frequently in social media texts, typically featured by informal language and spontaneous writing. Such occurrences can be linked to an abusive context, when they contribute to the expression of hatred and to the abusive effect, causing harm and offense. However, swearing is multifaceted and is often used in casual contexts, also with positive social functions. In this study, we explore the phenomenon of swearing in Twitter conversations, taking the possibility of predicting the abusiveness of a swear word in a tweet context as the main investigation perspective. We developed the Twitter English corpus SWAD (Swear Words Abusiveness Dataset), where abusive swearing is manually annotated at the word level. Our collection consists of 1,511 unique swear words from 1,320 tweets. We developed models to automatically predict abusive swearing, to provide an intrinsic evaluation of SWAD and confirm the robustness of the resource. We also present the results of a glass box ablation study in order to investigate which lexical, syntactic, and affective features are more informative towards the automatic prediction of the function of swearing.

The development of computational methods to detect abusive language in social media within variable and multilingual contexts has recently gained significant traction. The growing interest is confirmed by the large number of benchmark corpora for different languages developed in the latest years. However, abusive language behaviour is multifaceted and available datasets are featured by different topical focuses. This makes abusive language detection a domain-dependent task, and building a robust system to detect general abusive content a first challenge. Moreover, most resources are available for English, which makes detecting abusive language in low-resource languages a further challenge. We address both challenges by considering ten publicly available datasets across different domains and languages. A hybrid approach with deep learning and a multilingual lexicon to cross-domain and cross-lingual detection of abusive content is proposed and compared with other simpler models. We show that training a system on general abusive language datasets will produce a cross-domain robust system, which can be used to detect other more specific types of abusive content. We also found that using the domain-independent lexicon HurtLex is useful to transfer knowledge between domains and languages. In the cross-lingual experiment, we demonstrate the effectiveness of our jointlearning model also in out-domain scenarios.

This paper describes the participation of the #NonDicevoSulSerio team at SemEval2018-Task3, which focused on Irony Detection in English Tweets and was articulated in two tasks addressing the identification of irony at different levels of granularity. We participated in both tasks proposed: Task A is a classical binary classification task to determine whether a tweet is ironic or not, while Task B is a multiclass classification task devoted to distinguish different types of irony, where systems have to predict one out of four labels describing verbal irony by clash, other verbal irony, situational irony, and non-irony. We addressed both tasks by proposing a model built upon a well-engineered features set involving both syntactic and lexical features, and a wide range of affective-based features, covering different facets of sentiment and emotions. The use of new features for taking advantage of the affective information conveyed by emojis has been analyzed. On this line, we also tried to exploit the possible incongruity between sentiment expressed in the text and in the emojis included in a tweet. We used a Support Vector Machine classifier, and obtained promising results. We also carried on experiments in an unconstrained setting.