11763 research outputs found
Sort by
INTU-AI: Digitalization of Police Interrogation Supported by Artificial Intelligence
No full textArticle number - 10781Traditional police interrogation processes remain largely time-consuming and reliant on substantial human effort for both analysis and documentation. Intuition Artificial Intelligence (INTU-AI) is a Windows application designed to digitalize the administrative workflow associated with police interrogations, while enhancing procedural efficiency through the integration of AI-driven emotion recognition models. The system employs a multimodal approach that captures and analyzes emotional states using three primary vectors: Facial Expression Recognition (FER), Speech Emotion Recognition (SER), and Text-based Emotion Analysis (TEA). This triangulated methodology aims to identify emotional inconsistencies and detect potential suppression or concealment of affective responses by interviewees. INTU-AI serves as a decision-support tool rather than a replacement for human judgment. By automating bureaucratic tasks, it allows investigators to focus on critical aspects of the interrogation process. The system was validated in practical training sessions with inspectors and with a 12-question questionnaire. The results indicate a strong acceptance of the system in terms of its usability, existing functionalities, practical utility of the program, user experience, and open-ended qualitative responses.This work was partially supported by the Fundação para a Ciência e a Tecnologia, I.P. (FCT), under project reference UID/50008/2023 IT, and in the context of project UIDB/04524/2020, as well as through the Scientific Employment Stimulus (CEECINS/00051/2018). The authors also gratefully acknowledge the financial support from the European Union via the Portuguese Recovery and Resilience Plan, under the Call: 2021-C05i0101-02-agendas/alianças mobilizadoras para a reindustrialização—PRR, Proposal: C632482988-00467016
Fatores Determinantes da Aceitação da Inteligência Artificial no Marketing: O Papel Mediador da Digitalização e da Integração da IA nas Competências de Marketing
No full textA crescente integração da Inteligência Artificial (IA) nas práticas de marketing tem transformado a forma como as organizações tomam decisões, interagem com os consumidores e estruturam os seus processos internos. Esta evolução tecnológica exige uma compreensão mais aprofundada dos fatores que influenciam a aceitação da IA pelos profissionais de marketing.
A presente dissertação teve como objetivo identificar os determinantes da Aceitação da Inteligência Artificial (AIA) no marketing, analisando simultaneamente o papel mediador da Digitalização (DIG) e da Integração de Múltiplas Competências (M_IMC) — uma dimensão da multipotencialidade. Para tal, foi desenvolvido um modelo conceptual, testado através de um estudo quantitativo, descritivo e conclusivo, com base numa amostra não probabilística de conveniência composta por 208 profissionais de marketing a operar em Portugal. A análise dos dados foi conduzida com recurso ao software SmartPLS, com base em Modelação de Equações Estruturais.
Os resultados evidenciam que o Apoio Percebido da Gestão (APG) e a Digitalização (DIG) exercem um impacto significativo na AIA, enquanto a Facilidade de Uso Percebida (FUP) não parece ser relevante na aceitação da IA. A Integração da IA nas Competências de Marketing (IIACM) influencia positivamente a DIG e, as duas variáveis medeiam relação a relação entre a Integração de Múltiplas Competências e a aceitação da IA, permitindo concluir que a aceitação da IA, em parte, depende da predisposição do gestor de marketing para a Multipotencialidade em termos Integração de Múltiplas Competências (M_IMC) que se reflete na Integração da IA nas Competências de Marketing e consequentemente na digitalização. Contudo, a M_IMC não influencia diretamente a AIA.
A nível teórico, destaca-se o contributo inovador da construção e validação de uma nova escala para avaliar a IIACM, colaborando para colmatar uma lacuna existente na literatura. A nível prático, os resultados servem para fazer recomendações para o desenvolvimento de estratégias organizacionais que promovam a adoção da IA, valorizem perfis híbridos e multipotenciais, e incentivem uma cultura de digitalização e aprendizagem contínua. Apesar das limitações inerentes à amostragem e ao desenho transversal, esta investigação abre caminho a futuras pesquisas sobre a aceitação da IA em diferentes contextos organizacionais.The growing integration of Artificial Intelligence (AI) into marketing practices has been transforming the way organisations make decisions, interact with consumers, and structure their internal processes. This technological evolution requires a deeper understanding of the factors that influence AI acceptance among marketing professionals.
This dissertation aims to identify the determinants of Artificial Intelligence Acceptance (AIA) in marketing, while simultaneously analysing the mediating role of Digitalisation (DIG) and the Integration of Multiple Competencies (M_IMC) — a dimension of multipotentiality. A conceptual model was developed and tested through a quantitative, descriptive, and confirmatory study, based on a non-probability convenience sample comprising 208 marketing professionals operating in Portugal. Data analysis was conducted using SmartPLS software, applying Structural Equation Modelling (SEM).
The results reveal that Perceived Management Support (APG) and Digitalisation (DIG) have a significant impact on AIA, whereas Perceived Ease of Use (FUP) does not appear to influence AI acceptance. The Integration of AI into Marketing Competencies (IIACM) positively influences DIG, and both variables mediate the relationship between the Integration of Multiple Competencies and AI acceptance. These findings suggest that AIA partly depends on the marketing manager's predisposition toward Multipotentiality in terms of competency integration (M_IMC), which in turn is reflected in the integration of AI and digitalisation. However, M_IMC does not directly affect AIA.
Theoretically, this research contributes by developing and validating a novel scale for assessing IIACM, addressing a gap in the literature. Practically, the findings support recommendations for designing organisational strategies that foster AI adoption, promote hybrid and multipotential profiles, and encourage a culture of digitalisation and continuous learning.
Despite the limitations related to sampling and the cross-sectional design, this study paves the way for future research into AI acceptance across different organisational contexts
ESTUDO COMPARATIVO DA UTILIZAÇÃO DE LIVROS MULTIFORMATO EM FORMATO DIGITAL E FÍSICO POR PESSOAS COM DIFICULDADES INTELECTUAIS E DESENVOLVIMENTAIS
No full textContinuam a existir imensas limitações no acesso à informação por parte das pessoas com Dificuldades intelectuais e Desenvolvimentais (DID). Apesar de serem feitos imensos esforços no sentido de criar conteúdos por parte de várias instituições, nomeadamente por parte dos Centros de Recursos para a inclusão (CRI), existem limitações de divulgação projetual dos mesmos e capacidade de acesso pelo facto dos conteúdos serem tradicionalmente em formatos físicos e dessa forma limitados aos locais pelos quais se encontram distribuídos. Este estudo comparativo explora as preferências e efeitos do uso de livros em formato digital e físico em indivíduos com Dificuldades intelectuais e Desenvolvimentais (DID). Focando-se nas particularidades dos livros multiformato, especialmente ebooks com funcionalidades interativas, a investigação analisa como estas características influenciam a experiência de leitura, a retenção de informações e a ligação emocional com o conteúdo. Os resultados indicam que os indivíduos com DID tendem a preferir livros digitais devido à usabilidade que oferecem e a um maior dinamismo dos conteúdos, valorizando simultaneamente o papel dos formatos físicos. Destacam-se ainda, nos formatos digitais a acessibilidade e capacidade de personalização, sendo úteis em contextos específicos
Level up! How Gamed-Based Activities Transform Learning and Alleviate Stress in Institutionalized Elderly
No full textProceedings of the volume: Videogame Sciences and Arts
Conference name: 14th International Conference, Leiria, Portugal, December 5–6, 2024Part of the book series: Communications in Computer and Information Science (CCIS,volume 2324)Included in the following conference series: International Conference on Videogame Sciences and ArtsMental health issues are a critical concern for the elderly, as the inability to manage stress during stimulation activities can significantly impair their ability to accept and effectively learn new tasks, thereby affecting their performance in daily life activities. Serious games are increasingly recognized as valuable in the context of rehabilitation; however, there is a paucity of studies examining how elderly individuals manage stress and learn in regular practice using such games. In this study, 10 institutionalized elderly participants underwent 6 game-based stimulation sessions playing the serious games Ta!Ti! and Mexerico. Learning variables, including time and error rates, were assessed at baseline (T0), mid-point (T1), and the final session (T2), along with stress management indicators, specifically cortisol levels, at T0 and T2. The findings revealed that learning profiles improved throughout the program, with more pronounced gains observed initially. Additionally, stress levels decreased following each game-based session. The study identified significant relationships between stress management and learning profiles, suggesting that game-based activities can effectively enhance both learning outcomes and stress reduction in the elderly.This work has been made possible through the financial support of different key scholarships: the FI scholarship program of the Agència de Gestió d’Ajuts Universitaris i de Recerca de la Generalitat de Catalunya provided under the code “2023 FI-3 00107”; doctoral research scholarship program of the Foundation for Science and Technology of Portugal, with funding granted under the reference “2023. 02549.BDANA”; the competitive funding grants intended to finance the predoctoral hiring of research personnel in training by the agents of the Andalusian Knowledge System (code “24653”)
How Health Literacy impacts Polytechnic of Leiria Students?
No full textCENTERIS - International Conference on ENTERprise Information Systems / ProjMAN - International Conference on Project MANagement / HCist - International Conference on Health and Social Care Information Systems and TechnologiesIn 2021, aHealth Literacy(HL) evaluation among university students revealed notable limitations in HL. To assess the general HL of populations comprehensively, the European HLSurvey Questionnaire (HLS-EU-Q) was developed, encompassing 12 subdomains to provide a broad perspective on public health. In 2014, the questionnaire was adapted for use in Portugal, resulting in the HLS-EU-PT version, validated through a 16-question survey (HLS-EU-PT-Q16).Global HL andthreedomains’ indexes and levelswere determined, namely Healthcare (HC), Disease prevention (DP), and Health Promotion (HP). The HLSEU-Q16-PT assessment demonstrated satisfactory internal consistency, with 0.8834Cronbach's alpha coefficient.In this study, an online survey distributedbetween 2020-2021among Polytechnic of Leiria academia allowed data collection from various stakeholders, including 251 students, 109 professors, 15 researchers, and 55 other staff. From the430 responses,75 questions were analysed. The saved data wasthefocus of this work, regarding a thesis of the first edition of the master’s in data science to analysethe 251 surveyed studentsand their HL. The results revealed that thesestudents have lower HL index, and, in this case study,health areadegreeor school impactsHL.Acknowledgements This work has been funded by the Portuguese FCT-Fundação para a Ciência e a Tecnologia, I.P., within project UIDB/04524/2020. Pós-graduação em Gestão de Projetos -5ª Edição, 2021/2022, School of Technology and Management, Polytechnic of Leiria
Metodologias baseadas em Inteligência Artificial para a deteção de correio eletrónico não solicitado
No full textO problema da deteção de correio eletrónico não solicitado (Spam) é um que
esteve presente desde os primórdios da Internet levando a que, poucos anos após a
introdução desta, começasse o desenvolvimento de sistemas e algoritmos para detetar
e bloquear esta ameaça informática. Tentativas iniciais consistindo de moderação
humana com auxílio de relatórios pelas vítimas e de sistemas simples baseados no
bloqueio de certos termos ligados ao spam introduzidos numa blacklist manualmente
pelos seus criadores. No entanto, à medida que a deteção de spam evoluía, o spam
também evolui nas suas técnicas, desde táticas como a intencional introdução de erros
na escrita para confundir os sistemas de deteção rudimentares que não conseguiam
detetar corrupções de palavras contidas na sua blacklist, requerendo que os criadores
introduzam manualmente todas as possíveis corrupções eles próprios. Uma das mais
recentes inovações nesta área foi a de utilizar modelos de Inteligência Artificial, mais
especificamente, de Processamento de Linguagem Natural, para tentar criar um
modelo capaz de acompanhar indeterminadamente a evolução do spam, criando uma
solução permanente a este problema. Neste trabalho, é o objetivo treinar um modelo
NLP (NLP) baseado na arquitetura Sentence BERT (Bidirectional Representations
from Transformer), que possa servir de exemplo da capacidade e potencial desta no
combate ao spam, este modelo será treinado com um dataset de emails com mais de
90 mil emails sendo cerca de metade destes spam e a restante parte emails legítimos.
O modelo obtido demonstrou resultados positivos, tendo a versão final alcançado,
em combinação com um classificador Support Vector Classifier (SVC), métricas
que ultrapassaram os 98.5%, com o modelo sBERT a consumir cerca de 30 micro
segundos a vetorizar cada email.The issue of unsolicited electronic mail (spam) detection is one that has been
present since the begginings of the Internet leading to, just a few years after its’
introduction, the development of sistems and algorithms to detect and surpress
this threat. Initial efforts were primarily consistent of human intervention aided by
user reports, with rudimentary automated blocking systems based on a handcrafted
word blacklist appearing soon after. However, as spam detection evolved, so too
did spam itself evolve to counter these evolutions, with developments such as the
intentional miswriting of suspicious word so as to avoid detection by blacklisting,
requiring manual insertion of every possible corruption of a word into the blacklist
to counter it. One of the most recent development on the side of spam detection
has been the application Artificial Intelligence, more specifically, Natural Language
Processing Models (NLP) to the task, in an attempt to utilize its self improving
powers to create a system capable of trailing spams’ own evolution and thus, a
more permanent or, at least, longer lasting solution to the issue. In this work, the
goal is to train one such NLP model, based on the Sentence BERT (Bidirectional
Encoding Representations from Transformer) architecture, which may serve as an
example of the capacity and potential of this avenue in detecting spam, the model
in question will be trained with a dataset of various emails numbering just over 90
thousand with roughtly half of it consisting of spam and the remainder legitimate
emails. The resulting model displayed positive results, with the final result having,
in combination with an Support Vector Classifier (SVC), surpassed 98.5% in all
metrics, with the model taking an average of 30 micro seconds per email vectorized
Correlation between trace element concentrations in the blood of female hawksbill (Eretmochelys imbricata) and egg quality in nesting populations of São Tomé Island
No full textMetals and metalloids can pose a significant threat to sea turtles, as these contaminants tend to accumulate in their bodies over time, due to their long lifespans and varied feeding habits. São Tomé and Príncipe's archipelago hosts the last remaining rookery for hawksbill sea turtles (Eretmochelys imbricata) in the region. The study aimed to determine the levels of metals and metalloids accumulated by this population and to investigate their possible genotoxicity in nesting females' blood as well as potential effects on their eggs in terms of morphometric characteristics and the quality of their lipidic reserves, essential for embryo development. Higher levels of Hg were found to be correlated with increased “lobed-shaped nuclei” in erythrocytic count, suggesting genotoxicity effects in this population. Higher levels of Se were correlated with thicker and heavier eggshells, while Pb levels were associated with the reduction of the egg's diameter. Metal contamination in females' blood significantly affected yolk polar fatty acids. Significant negative correlations were found between general metal contamination (PLI) and saturated fatty acids (SFA), while positive correlations were observed for essential omega-6 fatty acids (n6), mostly influenced by Cu, Fe, and Hg concentrations. This suggests that these omega-6 fatty acids are being synthesized from SFA, potentially indicating stress response by metal exposure. The present results point to some potential alterations in the normal embryonic development of these turtle eggs, influenced by metal contamination, which should raise some concerns about the future of this critically endangered species and call for additional conservation efforts in the region.This study was supported by the Fundação para a Ciˆencia e a Tec nologia (FCT) through the Strategic Project granted to MARE (UID/04292/MARE–Centro de Ciˆencias do Mar e do Ambiente), the project granted to the Associated Laboratory ARNET (https://doi.org/10.5 4499/LA/P/0069/2020), the grant awarded to Inês Morão ao (https://doi.org/10.54499/PD/BD/150562/2019), and contracts to Tiago Simoes (https://doi.org/10.54499/2021.02559.CEECIND/CP1671/CT0001) and Sara Novais (https://doi.org/10.54499/CEECINST/00060/2021/CP2902/CT0007). We are grateful to all the members of the NGOs Pro grama Tato ˆ and Fundaçao ˜ Príncipe, as well as NGO research assistants (biologists Maria Branco and Yedda Oliveira), and Marco Santos and Manuel Armindo José, who worked on Rolas Islet during the sampling period and assisted us in the field under challenging conditions
Prevalence and Predictors of Long Covid in a Cohort of Brazilian Adults 12 Months After Acute Infection: A Cross‐Sectional Study
No full textArticle n. e70467Acknowledgements
The authors wish to thank all study participants and the healthcare professionals who contributed to the data collection.Introduction: Since the onset of the pandemic in early 2020, various reports have emerged regarding persistent symptoms associated with Covid‐19. Nevertheless, there is insufficient data on the persistence of symptoms over time. This study sought to estimate the prevalence of persistent symptoms 12 months after Covid‐19 infection and identify predictors of long Covid in
adults living in the State of Paraná, southern Brazil, according to the level of severity of Covid‐19 infection.
Method: An observational and cross‐sectional survey was conducted with Brazilian adults diagnosed with Covid‐19, as assessed from data available in two official Covid‐19 notification databases in Brazil, using telephone interviews. Descriptive statistics, tests of associations and simple and multiple binary logistic regression analysis were used to identify predictors of long Covid.
Results: In total, 1033 adults participated in the study. The overall prevalence of long Covid was 60.3% (n = 623). Prevalence was higher in women (67.7%), people aged between 50 and 59 years (65.8%) and in individuals who received treatment in an Intensive Care Unit (ICU) during the acute phase of Covid‐19 infection (74.4%, n = 241). The risk factors associated with a greater chance of developing long Covid were: female (OR 2.38; 95% CI 1.55; 3.66), living in the Brazilian northwest health macro‐region (OR 2.20; 95% CI 1.21; 4.00), presenting multimorbidity (OR 1.86; 95% CI 1.06; 3.28), having an average of six symptoms in the acute phase of Covid‐19 (OR 1.22; 95% CI 1.17; 1.28) and having received treatment in an ICU (OR 4.86; 95% CI 2.83; 8.35) and inpatient ward (OR 2.45; 95% CI 1.47; 4.09).
Conclusions: The results highlight the high prevalence of long Covid and support the formulation of health policies capable of minimising the consequences on the population, on the services offered by professionals and on health systems. Patient or Public Contribution: The study topic's importance was based on the patients' experiences in the author's previous research and the need to develop patient‐centred care.This study was funded by Ministério da Ciência, Tecnologia, Inovações e Comunicações (FNDCT/MCTIC), Ministério da Saúde (MS) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)—Processo n◦ 402882/2020-2. It was also supported by FCT—Fundação para a Ciência e a Tecnologia, I.P. (UID/05704/2023) and by the Scientific Employment Stimulus—Institutional Call—[https://doi.org/10.54499/CEECINST/00051/2018/CP1566/CT0012, accessed on 20 September 2025]
Monitoring Report on the 2024 Gender Equality, Non-Discrimination, and Inclusion Plan
No full textRevisão:
Cláudia Andreia Cunha Belém Toneca
Eva Patrícia da Silva Guilherme Menino
Inês Paulo Cordeiro Brasão
Jorge Alexandre Barroca de Sousa Varela
Maria Leopoldina Mendes Ribeiro de Sousa Alves
Rui Manuel da Fonseca Pinto
Sílvia Raquel Barros Pinto
Grafismo:
Laura Filipa Pedrosa FerreiraNo âmbito da implementação do Plano para a Igualdade de Género, Não Discriminação e Inclusão do Instituto Politécnico de Leiria, foi realizada a monitorização da sua execução do ano 2024. Este processo teve em consideração seis dimensões estratégicas para a comunidade académica, valorizando as ações desenvolvidas junto dos colaboradores e dos estudantes
Comunicação acessível no Centro Hospitalar de Leiria:Elaboração do guia de acolhimento do Serviço de Imagiologia
No full textO presente projeto de mestrado teve como objetivo a elaboração de um guia de acolhimento para o serviço de imagiologia do Centro Hospitalar de Leiria, com foco na importância da comunicação para a promoção da saúde. A comunicação, essencial na relação entre profissionais de saúde e utentes, é fundamental para garantir que as informações sejam transmitidas de forma clara e compreensível.
Neste contexto, a comunicação acessível emerge como uma abordagem necessária, para atender às diversas necessidades dos utentes, respeitando as particularidades e garantindo que todos tenham acesso igualitário às informações. O guia de acolhimento foi concebido com este princípio em mente, sendo estruturado de maneira a facilitar a compreensão de todos, independentemente de sua formação, idade ou condição de saúde.
A literacia em saúde é outro pilar deste projeto, uma vez que a capacidade dos indivíduos de interpretar, avaliar e utilizar informações sobre saúde é crucial para a tomada de decisões informadas. Ao fornecer um material acessível e de fácil entendimento, o guia visa capacitar os utentes, promovendo uma maior autonomia e participação no seu próprio cuidado.
A avaliação do guia foi realizada por meio de uma sondagem aplicado na sala de espera do serviço de imagiologia.
O presente estudo ressaltou a importância da acessibilidade e da literacia em saúde como ferramentas fundamentais para a promoção de um atendimento mais humano e eficaz, através do desenvolvimento do guia acessível em diferentes formatos: escrita fácil, texto aumentado, escrita pictográfica, áudio e língua gestual Portuguesa. A implementação de guias de acolhimento em outros serviços do hospital pode ampliar ainda mais estes benefícios