2024
Autores
Martins, D; Fernandes, C; Campos, MJ; Campos Ferreira, M;
Publicação
The International Journal of Information, Diversity, & Inclusion (IJIDI)
Abstract
2024
Autores
Gharahbagh, AA; Hajihashemi, V; Ferreira, MC; Machado, JM; Tavares, RS;
Publicação
Lecture Notes in Networks and Systems
Abstract
With the increasing popularity of social media platforms like Instagram, there is a growing need for effective methods to detect and analyze abnormal actions in user-generated content. Deep learning is part of a broader family of machine learning methods based on artificial neural networks with representation learning that can learn complex patterns. This article proposes a novel deep learning approach for detecting abnormal actions in social media clips, focusing on behavioural change analysis. The approach uses a combination of Deep Learning and textural, statistical, and edge features for semantic action detection in video clips. The local gradient of video frames, time difference, and Sobel and Canny edge detectors are among the operators used in the proposed method. The method was evaluated on a large dataset of Instagram and Telegram clips and demonstrated its effectiveness in detecting abnormal actions with about 86% of accuracy. The results demonstrate the applicability of deep learning-based systems in detecting abnormal actions in social media clips. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
2024
Autores
Hajihashemi, V; Gharahbagh, AA; Ferreira, MC; Machado, JM; Tavares, RS;
Publicação
Lecture Notes in Networks and Systems
Abstract
In recent years, social media platforms have become an essential source of information. Therefore, with their increasing popularity, there is a growing need for effective methods for detecting and analyzing their content in real time. Deep learning is a machine learning technique that teaches computers to understand complex patterns. Deep learning techniques are promising for analyzing acoustic signals from social media posts. In this article, a novel deep learning approach is proposed for socially contextualized event detection based on acoustic signals. The approach integrates the power of deep learning and meaningful features such as Mel frequency cepstral coefficients. To evaluate the effectiveness of the proposed method, it was applied to a real dataset collected from social protests in Iran. The results show that the proposed system can find a protester’s clip with an accuracy of approximately 82.57%. Thus, the proposed approach has the potential to significantly improve the accuracy of systems for filtering social media posts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
2024
Autores
Ferreira, MC; Veloso, M; Tavares, JMRS;
Publicação
Decision Support Systems XIV. Human-Centric Group Decision, Negotiation and Decision Support Systems for Societal Transitions - 10th International Conference on Decision Support System Technology, ICDSST 2024, Porto, Portugal, June 3-5, 2024, Proceedings
Abstract
Recent advancements in digital technology have significantly impacted healthcare, with the rise of chatbots as a promising avenue for healthcare services. These chatbots aim to provide prevention, diagnosis, and treatment services, thereby reducing the workload on medical professionals. Despite this trend, limited research has explored the variables influencing user experiences in the design of healthcare chatbots. While the impact of visual representation within chatbot systems is recognized, existing studies have primarily focused on efficiency and accuracy, neglecting graphical interfaces and non-verbal visual communication tools. This research aims to delve into user experience aspects of symptom checker chatbots, including identity design, interface layout, and visual communication mechanisms. Data was collected through a comprehensive questionnaire involving three distinct chatbots (Healthily, Mediktor and Adele – a self-developed solution) and underwent meticulous analysis, yielding valuable insights to aid the decision process when designing effective chatbots for symptom checking. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
2024
Autores
Gonçalves, HIT; Ferreira, MC; Campos, MJ; Fernandes, CS;
Publicação
International Journal of Orthopaedic and Trauma Nursing
Abstract
Objective: This study aims to present the developmental stages of a Mobile App Prototype designed to enhance patient participation in the rehabilitation process after hip replacement. Methods: To ensure effective interaction between the system and the end user, a User-Centered Design methodology was followed, encompassing three phases: Requirements gathering, Prototyping, and Evaluation. Usability tests were conducted to assess the usability of the developed system. Results: The RehabApp for mobile devices was created, and the testing results were positive. Users expressed satisfaction with the outcome, deeming it a valuable tool for their recovery. This outcome demonstrates the high receptiveness of these technologies in the healthcare sector, making it a project that can readily be expanded into other areas of rehabilitation. Conclusion: This study demonstrated the potential of the RehabApp in the rehabilitation process after hip replacement surgery. This accomplishment was realized by ensuring the active participation of patients, potential users, and healthcare professionals throughout the app's development. Innovation: The RehabApp is a mobile application to provide users with all the necessary knowledge, enabling them to undergo a smoother and safer rehabilitation. Feedback from both patients and healthcare professionals played a crucial role in refining the app's features and addressing usability concerns. © 2024 The Authors
2024
Autores
Vaz, TG; Oliveira, BB; Brandão, L;
Publicação
Applied Energy
Abstract
In the energy production sector, increasing the quantity and efficiency of renewable energies, such as hydropower plants, is crucial to mitigate climate change. This paper proposes a new and flexible model for optimising operational decisions in watershed systems with interconnected dams. We propose a systematic representation of watersheds by a network of different connection points, which is the basis for an efficient Mixed-Integer Linear Programming model. The model is designed to be adaptable to different connections between dams in both main and tributary rivers. It supports decisions on power generation, pumping and water discharge, maximising profit, and considering realistic constraints on water use and factors such as future energy prices and weather conditions. A relax-and-fix heuristic is proposed to solve the model, along with two heuristic variants to accommodate different watershed structures and sizes. Methodological tests with simulated instances validate their performance, with both variants achieving results within 1% of the optimal solution faster than the model for the tested instances. To evaluate the performance of the approaches in a real-world scenario, we analyse the case study of the Cávado watershed (Portugal), providing relevant insights for managing dam operations. The model generally follows the actual decisions made in typical situations and flood scenarios. However, in the case of droughts, it tends to be more conservative, saving water unless necessary or profitable. The model can be used in a decision-support system to provide decision-makers with an integrated view of the entire watershed and optimised solutions to the operational problem at hand. © 2024 The Author(s)
The access to the final selection minute is only available to applicants.
Please check the confirmation e-mail of your application to obtain the access code.