Abstract
Medical imaging classification improves patient prognoses by providing information on disease assessment, staging, and treatment response. The high demand for medical imaging acquisition requires the development of effective classification methodologies, occupying deep learning technologies, the pool position for this task. However, the major drawback of such techniques relies on their black-box nature which has delayed their use in real-world scenarios. Interpretability methodologies have emerged as a solution for this problem due to their capacity to translate black-box models into clinical understandable information. The most promising interpretability methodologies are concept-based techniques that can understand the predictions of a deep neural network through user-specified concepts. Concept activation regions and concept activation vectors are concept-based implementations that provide global explanations for the prediction of neural networks. The explanations provided allow the identification of the relationships that the network learned and can be used to identify possible errors during training. In this work, concept activation vectors and concept activation regions are used to identify flaws in neural network training and how this weakness can be mitigated in a human-in-the-loop process automatically improving the performance and trustworthiness of the classifier. To reach such a goal, three phases have been defined: training baseline classifiers, applying the concept-based interpretability, and implementing a human-in-the-loop approach to improve classifier performance. Four medical imaging datasets of different modalities are included in this study to prove the generality of the proposed method. The results identified concepts in each dataset that presented flaws in the classifier training and consequently, the human-in-the-loop approach validated by a team of 2 clinicians team achieved a statistically significant improvement.

Abstract
Most forecasting methods use recent past observations (lags) to model the future values of univariate time series. Selecting an adequate number of lags is important for training accurate forecasting models. Several approaches and heuristics have been devised to solve this task. However, there is no consensus about what the best approach is. Besides, lag selection procedures have been developed based on local models and classical forecasting techniques such as ARIMA. We bridge this gap in the literature by carrying out an extensive empirical analysis of different lag selection methods. We focus on deep learning methods trained in a global approach, i.e., on datasets comprising multiple univariate time series. Specifically, we use NHITS, a recently proposed architecture that has shown competitive forecasting performance. The experiments were carried out using three benchmark databases that contain a total of 2411 univariate time series. The results indicate that the lag size is a relevant parameter for accurate forecasts. In particular, excessively small or excessively large lag sizes have a considerable negative impact on forecasting performance. Cross-validation approaches show the best performance for lag selection, but this performance is comparable with simple heuristics. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Abstract
This work presents a specialized tabu search algorithm applied to the problem of electric power distribution systems primary feeders' reconfiguration. The specialization is related to two fundamental aspects of the tabu search algorithm. The first proposal eliminates the concept of a list of prohibited attributes and the aspiration criterion, but also avoids the possibility of revisiting a candidate solution so that cycling is avoided by maintaining a tabu list with all previously visited solutions. The second proposal is the possibility of restarting the search from the incumbent solution while avoiding paths that can be formed by revisiting candidate solutions. A new strategy based on Prim's algorithm generates a high-quality initial solution for the problem. Tests are conducted using the 33-, 84-, 118-, 136-, and 415-node test systems. The results demonstrate the effectiveness of the proposal for solving the reconfiguration problem since the best-known solution for each system is achieved within highly efficient execution times.

Abstract

This document is intended to present a benchmark of multiple good practices in the context of internationalization studies, particularly focused on digital electronics and programmable devices, yet is not limited to them. This paper will start with a comprehensive paper desk analysis together with an in-depth research process that should lead to the selection of innovative tools applied to digital systems. International initiatives are oriented towards increasing the quality of higher education by motivating teachers of STEM disciplines to use a multidisciplinary approach and teach with the massive support of technologies like Classroom, MS-Teams, Blackboard, etc. The central goal is to suggest and recommend a model for integrating intermediate and advanced digital electronics subjects (e.g., FPGA, microcontrollers, etc.) and ICT in international teaching approaches such as Collaborative Online International Learning (COIL), Project-based Learning (PBL) and Real Remote Labs (RRL). This is the approach sought by the European Project DECEL.

Abstract
The increase in renewable generation of a distributed nature has brought significant new challenges to power and energy system management and operation. Self-consumption in buildings is widespread, and with it rises the need for novel, adaptive and intelligent building energy management systems. Although there is already extensive research and development work regarding building energy management solutions, the capabilities for adaptation and contextualization of decisions are still limited. Consequently, this paper proposes a novel contextual rule-based system for energy management in buildings, which incorporates a contextual dimension that enables the adaptability of the system according to diverse contextual situations and the presence of multiple users with different preferences. Results of a case study based on real data show that the contextualization of the energy management process can maintain energy costs as low as possible, while respecting user preferences and guaranteeing their comfort.

Abstract
The human population with disability is rapidly expanding, more than 15% of people worldwide suffer from a disability and, despite the availability of accessibility guidelines, the websites are still inaccessible. Moreover, professionals with knowledge of accessibility and design abilities are hard to come by. Therefore, the current paper addresses the introduction of accessibility to the Software Engineering students through AccessCademy, a gamification-based tool, in a fun way. The activity is delivered via a Web-based learning environment, that presents bad accessibility scenarios or failures based on the Web Content Accessibility Guidelines (WCAG), and then encourages the students to solve them. Furthermore, a case study will be presented that evaluated the learning effectiveness of the tool in the context of a university course. The results demonstrated the potential of AccessCademy which offers students a fun and engaging way to learn about accessibility, to understand the importance of accessible design with WCAG and gain accessible design skills as well. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.