Browsing by Issue Date, starting with "2026"
Now showing 1 - 10 of 18
Results Per Page
Sort Options
- Optimization, Learning Algorithms and Applications: 5th International Conference, OL2A 2025 - Part 1Publication . Pereira, Ana I.; Fernandes, Florbela P.; Coelho, João Paulo; Teixeira, João Paulo; Lima, José; Pacheco, Maria F.; Oneto, Luca; Lopes, Rui PedroThe volumes CCIS 2617 and 2618 contain the refereed proceedings of the V International Conference on Optimization, Learning Algorithms and Applications (OL2A 2025), a hybrid event held on April 28–30. OL2A provided a space for the research community on optimization and learning to get together and share the latest developments, trends and techniques as well as develop new paths and collaborations. OL2A had the participation of more than two hundred participants in an online and face-to-face environment throughout three days, discussing topics associated with areas such as optimization and learning and state-of-the-art applications related to multi-objective optimization, optimization for machine learning, robotics, health informatics, data analysis, optimization and learning under uncertainty and the 4th industrial revolution. Four special sessions were organized under the topics Artificial Intelligence in Healthcare and Medicine, Optimization in the SDG Context, Optimization in Control Systems Design, and Machine Learning and Artificial Intelligence in Robotics. The event had 38 accepted papers. All papers were carefully reviewed and selected from the 92 received submissions, with each paper receiving three double-blind reviews on average. All the reviews were carefully carried out by a scientific committee of 115 researchers from twenty-one countries.
- Optimization, Learning Algorithms and Applications: 5th International Conference - Part 2Publication . Pereira, Ana I.; Fernandes, Florbela P.; Coelho, João Paulo; Teixeira, João Paulo; Lima, José; Pacheco, Maria F.; Oneto, Luca; Lopes, Rui PedroThe volumes CCIS 2617 and 2618 contain the refereed proceedings of the V International Conference on Optimization, Learning Algorithms and Applications (OL2A 2025), a hybrid event held on April 28–30. OL2A provided a space for the research community on optimization and learning to get together and share the latest developments, trends and techniques as well as develop new paths and collaborations. OL2A had the participation of more than two hundred participants in an online and face-to-face environment throughout three days, discussing topics associated with areas such as optimization and learning and state-of-the-art applications related to multi-objective optimization, optimization for machine learning, robotics, health informatics, data analysis, optimization and learning under uncertainty and the 4th industrial revolution. Four special sessions were organized under the topics Artificial Intelligence in Healthcare and Medicine, Optimization in the SDG Context, Optimization in Control Systems Design, and Machine Learning and Artificial Intelligence in Robotics. The event had 38 accepted papers. All papers were carefully reviewed and selected from the 92 received submissions, with each paper receiving three double-blind reviews on average. All the reviews were carefully carried out by a scientific committee of 115 researchers from twenty-one countries.
- Forecasting COVID-19 in european countries using long short-term memoryPublication . Carvalho, Kathleen; Teixeira, Rita de Almeida; Reis, Luis Paulo; Teixeira, João PauloEffective time series forecasts are increasingly important in supporting judgment in various decisions. Various prediction models are available to support these projections based on how each area provides a diverse set of data with variable behavior. Artificial neural networks (ANNs) significantly contribute to medical research since using predictive ideas allows for the study of disease progression in the future, as well as the behavior of other variables. This study implemented the proposed model based on Long Short-Term Memory (LSTM) to forecast COVID-19 daily new cases, deaths, and ICU patients. The methodology uses quantitative and qualitative data from six European countries: Austria, France, Germany, Italy, Portugal, and Spain to predict the last 242 days of the COVID-19 pandemic. The dataset uses the healthcare parameters of the number of daily new cases, deaths, ICU patients, and mitigation procedures, such as the percentage of the population fully vaccinated, the mandatory use of masks, and the lockdown. Two approaches were used to evaluate the model’s performance: the mean absolute error (MAE) and the mean square error (MSE). The results demonstrate that the LSTM model efficiently captures general trends in COVID-19 metrics but shows limitations when predicting data with low values or high variability, such as daily deaths. The model reported the lowest errors for Spain and Portugal, while France and Germany exhibited higher error rates due to differences in data reporting and pandemic dynamics. These findings highlight the importance of contextualizing predictive models based on specific regional characteristics.
- Influence of habits and comorbidities on liver diseasePublication . Leite, Gabriel A.; Azevedo, Beatriz Flamia; Pacheco, Maria F.; Fernandes, Florbela P.; Pereira, Ana I.The prevalence of hepatocellular carcinoma is expected to continue increasing worldwide, and its difficulty in early detection highlights the need for advanced monitoring technologies. As the disease progresses, it has a serious impact on patients’ health, and in severe cases, liver transplantation becomes the only viable solution, reinforcing its importance as a global health problem. This study proposes the use of different artificial intelligence methods to compare and understand them related to liver disease. Well-known algorithms such as Random Forest and Multi-Layer Perceptron were tested, as well as ensemble methods that exploit different modeling structures. The results showed that AdaBoost, Random Forest, and Gradient Boosting performed best with Area Under the Curve of 0.89, 0.86, and 0.84 respectively. To analyze their influence on clinical results, the best-performing model was reapplied only to the non-biochemical features that compose the dataset. The results indicate that portal vein thrombosis, diabetes, and hypertension are the most influential variables, with contributions of 29.48%, 20.50%, and 16.60%, respectively.
- Comparing RL policies for robotic pusherPublication . Bonjour, Pedro; Lopes, Rui PedroReinforcement learning (RL) has been consolidated as a promising approach to optimizing robotic tasks, allowing the improvement of performance and energy efficiency. This study investigates the effectiveness of five RL algorithms in the Pusher environment. Advantage Actor-Critic (A2C), Proximal Policy Optimization (PPO), Deep Deterministic Policy Gradient (DDPG), Soft Actor-Critic (SAC) and Twin Delayed Deep Deterministic Policy Gradient (TD3). We evaluated training time, computational efficiency, and reward values to identify the most balanced solution between accuracy and energy consumption. The results indicate that the PPO offers the best compromise between performance and efficiency, with reduced training time and stability in learning. SAC achieves the best rewards but requires more training time, while A2C faces difficulties in continuous spaces. DDPG and TD3, despite t he good results, have high computational consumption, which limits their viability for real-time industrial applications. These findings highlight the importance of considering energy efficiency when choosing RL algorithms for robotic applications. As a future direction, we propose the implementation of these algorithms in a real-world environment, as well as the exploration of hybrid approaches that combine different strategies to improve accuracy and minimize energy consumption.
- An openmodelica package for BELBICPublication . Coelho, João Paulo; Coelho, J. A. B.; Braz-César, ManuelThis paper presents the development and implementation of a new package for OpenModelica that integrates the Brain Emotional Learning Based Intelligent Controller (BELBIC) approach into control system simulations. BELBIC, inspired by neurobiological models of emotional learning, has demonstrated effectiveness in handling complex, nonlinear, and adaptive control problems. The proposed package provides a modular and user-friendly framework for integrating BELBIC enabling researchers and engineers to design, simulate, and analyze intelligent control strategies within an open-source environment. Key features of the package include customizable emotional response parameters and compatibility with existing Modelica libraries. To validate the package, a set of examples are included which demonstrates its application to the control of common dynamic systems.
- A deep learning approach for average height estimation in oak colony using rgb imagesPublication . Britto, Raphael Duarte; Mendes, João; Grilo, Vinicius; Castro, João Paulo; Santos, Murillo Ferreira dos; Castro, Marina; Pereira, Ana I.; Lima, JoséMany strategies have been developed to monitor the volume of volume of Above Ground Biomass (AGB) in forest areas as a fundamental step for managing carbon concentration. This study explores the use of use of Light Detection and Ranging (LiDAR) data obtained through Unmanned Aerial Vehicles (UAVs) to estimate height values in a vegetation colony composed of oaks (Quercus pyrenaica Willd.) in northern Portugal. The extraction of pertinent information from LiDAR data was facilitated by using the LAStools extension within the Quantum Geographic Information System (QGIS) software framework. The generated raster and image information were used to calculate the height values of the vegetation. Following this extraction, the information was meticulously organized into datasets, which were then employed in Deep Learning (DL) algorithms. The VGG16 model was selected as the underlying framework for the present study. Height predictions were made using dimensions of 16× 16, 32× 32, and 64 × 64 pixels for the Red, Green and Blue (RGB) images. The data was estimated and compared using both the standard format of the VGG16 model and a superficially adapted version of its convolution layers. The algorithm’s efficacy was validated by comparing the forecast results with the data obtained from QGIS, which revealed minimal discrepancies. It was observed that using 64× 64 pixel scale images yielded enhanced accuracy, resulting in reduced values for the Mean Absolute Error (MAE). The study demonstrates the viability of applying DL techniques to accurately capture information about a forest area using RGB images.
- Performance benchmarking of or-tools methods for capacitated vehicle routing problems with time windowsPublication . Sena, Inês; Ribeiro, Tiago B.; Silva, Adriano S.; Fernandes, Florbela P.; Costa, Lino A.; Pereira, Ana I.The Capacitated Vehicle Routing Problem with Time Windows (CVRPTW) is a significant challenge in combinatorial optimization, with extensive practical applications in logistics and transportation. This study aims to conduct a comparative analysis of the various methods available in OR-Tools for solving the CVRPTW across datasets of different sizes and types using the Solomon and the Gehring and Homberger benchmarks. The analysis provided insights into the relative strengths of each method, with a primary focus on Guided Local Search (GLS) and Tabu Search (TS), showing consistent performance and adaptability to different dataset characteristics. The results indicate that GLS is the most robust method overall, and TS can outperform it in specific scenarios. In conclusion, this study offers insights for selecting the most effective method to solve vehicle routing problems based on the characteristics and scale of the problem.
- Interconnection between lifestyle, health, and academic outcomes: an analysis on study habits and well-beingPublication . Azevedo, Beatriz Flamia; Bezerra, Ana J.B.; Sirmakessis, Spiros; Pereira, Ana I.Balancing academic demands with personal and professional responsibilities has become an increasingly challenging task, making it difficult to maintain well-being and potentially leading to serious health problems. The stress resulting from these multiple daily tasks, combined with the pressure to perform at high academic levels, directly impacts students’ mental and emotional health, significantly compromising their quality of life. In this work, statistical and clustering techniques are employed to analyze the dataset “Daily Lifestyle and Academic Performance of Students”. The objective of this work is to explore the relationship between students’ daily habits, level of stress, and the impact on academic performance. The results point out that many students have difficulty managing time and maintaining well-being (low-stress levels) with high academic performance since, according to the results, the higher the academic outcome, the higher the student’s stress level.
- Automated preprocessing of olive leaf images for cultivar classification using YOLO11Publication . Mendes, João; Lima, José; Rodrigues, Nuno; Pereira, Ana I.Olive cultivation is a pillar of Mediterranean agriculture, deeply rooted in both tradition and economic importance. This paper presents a novel two-phase methodology for the automated preprocessing of olive leaf images to facilitate accurate cultivar classification. Leveraging the state-of-the-art YOLO11 framework, two models (YOLO11n and YOLO11s) were employed for detection and segmentation tasks. A comprehensive dataset, combining in-situ captured images with publicly available data, was meticulously annotated using both manual and semi-automatic processes. The detection model identifies individual olive leaves, while the segmentation model isolates the leaves by replacing the background with a uniform white, thereby simulating laboratory conditions. Experimental results demonstrate that YOLO11n outperforms YOLO11s in terms of mean Average Precision and F1-score, confirming the feasibility of deploying the system on mobile devices for real-time, in-field classification.