The rapid proliferation of hate speech on social media poses significant challenges to maintaining a safe and inclusive digital environment. This paper presents a comprehensive review of automatic hate speech detection methods, with a particular focus on the evolution of approaches from traditional machine learning and deep learning models to the more advanced Transformer-based architectures. We systematically analyze over 100 studies, comparing the effectiveness, computational requirements, and applicability of various techniques, including Support Vector Machines, Long Short-Term Memory networks, Convolutional Neural Networks, and Transformer models like BERT and its multilingual variants. The review also explores the datasets, languages, and sources used for hate speech detection, noting the predominance of English-focused research while highlighting emerging efforts in low-resource languages and cross-lingual detection using multilingual Transformers. Additionally, we discuss the role of generative and multi-task learning models as promising avenues for future development. While Transformer-based models consistently achieve state-of-the-art performance, this review underscores the trade-offs between performance and computational cost, emphasizing the need for context-specific solutions. Key challenges such as algorithmic bias, data scarcity, and the need for more standardized benchmarks are also identified. This review provides crucial insights for advancing the field of hate speech detection and shaping future research directions.

The rapid rise of social media has brought about new ways of digital communication, along with a worrying increase in online hate speech (HS), which, in turn, has led researchers to develop several Natural Language Processing methods for its detection. Although significant strides have been made in automating HS detection, research focusing on the European Portuguese language remains scarce (as it happens in several under-resourced languages). To address this gap, we explore the efficacy of various transfer learning models, which have been shown in the literature to have better performance for this task than other Deep Learning models. We employ BERT-like models pre-trained on Portuguese text, such as BERTimbau and mDeBERTa, as well as GPT, Gemini and Mistral generative models, for the detection of HS within Portuguese online discourse. Our study relies on two annotated corpora of YouTube comments and tweets, both annotated as HS and non-HS. Our findings show that the best model for the YouTube corpus was a variant of BERTimbau retrained with European Portuguese tweets and fine-tuned for the HS task, with an F-score of 87.1% for the positive class, outperforming the baseline models by more than 20% and with a 1.8% increase compared with base BERTimbau. The best model for the Twitter corpus was GPT-3.5, with an F-score of 50.2% for the positive class. We also assess the impact of using in-domain and mixed-domain training sets, as well as the impact of providing context in generative model prompts on their performance.

Two sentences can be related in many different ways. Distinct tasks in natural language processing aim to identify different semantic relations between sentences. We developed several models for natural language inference and semantic textual similarity for the Portuguese language. We took advantage of pre-trained models (BERT); additionally, we studied the roles of lexical features. We tested our models in several datasets—ASSIN, SICK-BR and ASSIN2—and the best results were usually achieved with ptBERT-Large, trained in a Brazilian corpus and tuned in the latter datasets. Besides obtaining state-of-the-art results, this is, to the best of our knowledge, the most all-inclusive study about natural language inference and semantic textual similarity for the Portuguese language.

Several cases of autistic children successfully interacting with virtual assistants such as Siri or Cortana have been recently reported. In this demo we describe ChatWoz, an application that can be used as a Wizard of Oz, to collect real data for dialogue systems, but also to allow children to interact with their caregivers through it, as it is based on a virtual agent. ChatWoz is composed of an interface controlled by the caregiver, which establishes what the agent will utter, in a synthesised voice. Several elements of the interface can be controlled, such as the agent's face emotions. In this paper we focus on the scenario of child-caregiver interaction and detail the features implemented in order to couple with it.