A5023246066- Topic Modeling
- Natural Language Processing Techniques
- Sentiment Analysis and Opinion Mining
- Hate Speech and Cyberbullying Detection
- Advanced Text Analysis Techniques
- Social Media and Politics
- Speech and dialogue systems
- Misinformation and Its Impacts
- Semantic Web and Ontologies
- Education and Public Policy
- Spam and Phishing Detection
- Social and Political Issues
- Biomedical Text Mining and Ontologies
- Mental Health via Writing
- Internet Traffic Analysis and Secure E-voting
- Gender, Feminism, and Media
- Soil Management and Crop Yield
- Language, Metaphor, and Cognition
- Linguistics and Language Studies
- Freedom of Expression and Defamation
- Geographic Information Systems Studies
- Text Readability and Simplification
- Public Health in Brazil
- Data Quality and Management
- Business and Management Studies
Instituto Politécnico de Lisboa
2002-2024
University of Lisbon
2001-2024
Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento
2012-2024
Universidade Europeia
2016-2020
Universidade Nova de Lisboa
2019
Universidade Federal do Tocantins
2018
Instituto Superior Técnico
2002
We introduce a deep neural network for automated sarcasm detection.Recent work has emphasized the need models to capitalize on contextual features, beyond lexical and syntactic cues present in utterances.For example, different speakers will tend employ regarding subjects and, thus, detection ought encode such speaker information.Current methods have achieved this by way of laborious feature engineering.By contrast, we propose automatically learn then exploit user embeddings, be used concert...
We investigate the accuracy of a set surface patterns in identifying ironic sentences comments submitted by users to an on-line newspaper. The initial focus is on irony containing positive predicates since these are more exposed irony, making their true polarity harder recognize. show that it possible find with relatively high precision (from 45% 85%) exploring certain oral or gestural clues user comments, such as emoticons, onomatopoeic expressions for laughter, heavy punctuation marks,...
We propose and evaluate a method for automatically creating reference corpus training text classification procedures mining political opinions in user-generated content. The process starts by compiling collection of highly opinionated comments posted users on an on-line newspaper. Then, we define use set manually-crafted high-precision rules supported large sentiment-lexicon order to identify sentences each comment expressing about entities. Finally, the found are propagated remainder...
This paper investigates the pervasive issue of hate speech within Twitter/X Portuguese network conversations, offering a multifaceted analysis its characteristics. study utilizes mixed-method approach, combining several methodologies (triad census and participation shifts) over interaction between users. Qualitative manual content annotation was applied to dataset dissect different patterns on platform. Key findings reveal that number users followed by an individual potentially reads is...
Abstract This paper addresses the specificities of online hate speech against Afro-descendant, Roma, and LGBTQ+ communities in Portugal. The research is based on analysis CO-HATE, a corpus composed 20,590 YouTube comments, which were manually annotated following detailed guidelines created for that purpose. We applied methods from linguistics to assess prevalence overt covert speech, counter-speech, offensive considering different grounds discrimination, investigate main linguistic...
This paper describes the main characteristics of SentiLex-PT, a sentiment lexicon designed for extraction and opinion about human entities in Portuguese texts. The potential this resource is illustrated on its application to two types corpora, SentiCorpus-PT, social media corpus, consisting user comments news articles, literary piece early twentieth century, Poor (Os Pobres), by Raul Brandão. data were processed UNITEX, natural language processing system based dictionaries grammars.
Abstract New and flexible educational paradigms, based on creative, innovative open‐minded competences, are required in the development of curricula design, working as an essential skill toolkit for future designers, particularly higher education. This study aims to explore how learning outcomes, usually expressed by knowledge, skills, abilities, attitudes competences expected be achieved students a result experience, defined formulated design programmes Portugal. The investigation relies...
Video is a very rich medium that becoming increasingly dominant. A massive amount of video information available, but difficult to access if not adequately indexed: challenging task accomplish. We describe Information Retrieval system, under development, operates on database composed subtitled documents. The simultaneous analysis video, subtitles and audio streams performed in order index, visualize retrieve excerpts documents share certain emotional or semantic property.