Iscte Repository
University Institute of Lisbon
ISCTE institutional repository aims to preserve, disseminate and give access to its intellectual production in digital format. By collecting the academic and scientific publications Iscte repository contributes to increase their visibility and the research work impact at national and international level.

Search
Recent Submissions
Item type: Item , Access status: Open Access , Europe’s cognitive security: Generative AI, hybrid conflict and democratic resilience(2026) Marsili, M.The spread of generative AI, hybrid conflict and foreign information manipulation is reshaping the European security environment. The European Union is responding by moving beyond a narrow counter-disinformation approach towards a broader concept of cognitive security. Yet major gaps remain in enforcement, strategic coherence and democratic resilience.Item type: Item , Access status: Open Access , Bypassing the nuances of Portuguese covert hate speech through contextual analysis(2025) Ramos, G.; Batista, F.; Ribeiro, R.; Fialho, P.; Moro, S.; Fonseca, A.; Guerra, R.; Carvalho, P.; Marques, C.; Silva, C.; Santos, Manuel Filipe; Machado, José; Novais, Paulo; Cortez, Paulo; Moreira, Pedro MiguelDetecting and addressing covert hate speech poses significant challenges in online platforms where discriminatory messages are often disguised within seemingly innocuous content. In this study, we investigate the effectiveness of contextual analysis in bypassing the nuances of covert hate speech. Our research explores the impact of prompt engineering and context addition on the classification of overt and covert hate speech across diverse target groups, including Roma, migrants, LGBTQ+, and individuals of African descent. Through experimental trials using generative models like GPT-3.5 and GPT-4, our findings reveal that the addition of context, not only improves the overall performance of the models in general hate speech (from 75.0% to 79.6% F1 score, for the positive class), but also significantly improves the classification of covert hate speech, increasing True Positives by 21.64% (absolute) compared to the 6.5% in overt hate speech. Despite these improvements, the addition of context also increased the number of False Positives, indicating that a further refinement is needed for this contextual analysis. Moreover, target group analysis demonstrates a correlation between the prevalence of covert hate speech and model performance.Item type: Item , Access status: Open Access , Segmentation model for judgments of the Portuguese supreme court of justice(2025) Zanatti, M.; Ribeiro, R.; Pinto, H. S.; Santos, Manuel Filipe; Machado, José; Novais, Paulo; Cortez, Paulo; Moreira, Pedro MiguelLegal document segmentation is a critical task in the field of natural language processing (NLP), enabling efficient analysis, retrieval, and understanding of legal content. Despite its importance, research in this area for European Portuguese has been limited. To address this gap, we present a novel approach to automate the segmentation of legal judgments from the Portuguese Supreme Court of Justice into distinct sections. Leveraging a Bi-LSTM-CRF model, we developed a dataset and achieved significant results, including an accuracy of 0.9997, precision of 0.9986, recall of 0.996, and F1-Score of 0.9973. Our methodology and experimental results demonstrate the effectiveness and potential applications of our approach for the European Portuguese language.
