Abstract
With the outstanding predictive performance of Convolutional Neural Networks on different tasks and their widespread use in real-world scenarios, it is essential to understand and trust these black-box models. While most of the literature focuses on post-model methods, we propose a novel in-model joint architecture, composed by an explainer and a classifier. This architecture outputs not only a class label, but also a visual explanation of such decision, without the need for additional labelled data to train the explainer besides the image class. The model is trained end-to-end, with the classifier taking as input an image and the explainer's resulting explanation, thus allowing for the classifier to focus on the relevant areas of such explanation. Moreover, this approach can be employed with any classifier, provided that the necessary connections to the explainer are made. We also propose a three-phase training process and two alternative custom loss functions that regularise the produced explanations and encourage desired properties, such as sparsity and spatial contiguity. The architecture was validated in two datasets (a subset of ImageNet and a cervical cancer dataset) and the obtained results show that it is able to produce meaningful image- and class-dependent visual explanations, without direct supervision, aligned with intuitive visual features associated with the data. Quantitative assessment of explanation quality was conducted through iterative perturbation of the input image according to the explanation heatmaps. The impact on classification performance is studied in terms of average function value and AOPC (Area Over the MoRF (Most Relevant First) Curve). For further evaluation, we propose POMPOM (Percentage of Meaningful Pixels Outside the Mask) as another measurable criteria of explanation goodness. These analyses showed that the proposed method outperformed state-of-the-art post-model methods, such as LRP (Layer-wise Relevance Propagation).

Abstract
This article proposes a robotic system that aims to support the elderly, to comply with the medication regimen to which they are subject. The robot uses its locomotion system to move to the elderly and through computer vision detects the packaging of the medicine and identifies the person who should take it at the correct time. For the accomplishment of the task, an application was developed supported by a database with information about the elderly, the medicines that they have prescribed and the respective timetable of taking. The experimental work was done with the robot NAO, using development tools like MySQL, Python, and OpenCV. The elderly facial identification and the detection of medicine packing are performed through computer vision algorithms that process the images acquired by the robot's camera. Experiments were carried out to evaluate the performance of object recognition, facial detection, and facial recognition algorithms, using public databases. The tests made it possible to obtain qualitative metrics about the algorithms' performance. A proof of concept experiment was conducted in a simple scenario that recreates the environment of a dwelling with seniors who are assisted by the robot in the taking of medicines.

Abstract
COVID-19 has caused dramatic effects on the world economy, business activities, and people. But digitization is also helping many companies to adapt and overcome the current situation caused by COVID-19. The growth in the use of technology in the daily lives of people and companies to face this exceptional situation is an evidence of the digital acceleration process. This exploratory study analyzes the impact of digital transformation processes in three business areas: labor and social relations, marketing and sales, and technology. The impact of digitalization is expected to be transversal to each area and will encourage the emergence of new digital products and services based on the principle of flexibility. Additionally, new ways of working will foster the demand for new talent regardless of people's geographical location. Moreover, cybersecurity and privacy will become two key elements that will support the integrated development of the Internet of Things technology solutions, artificial intelligence, big data, and robotics. IEEE

Abstract
Research data management is the basis for making data more Findable, Accessible, Interoperable and Reusable. In this context, little attention is given to research data in image format. This article presents the preliminary results of a study on the habits related to the management of images in research. We collected 107 answers from researchers using a questionnaire. These researchers were PhD students, fellows and university professors from Life and Health Sciences, Exact Sciences and Engineering, Natural and Environmental Sciences and Social Sciences and Humanities. This study shows that 83.2% of researcher use images as research data, however, its use is generally not accompanied by a guidance document such as a research data management plan. These results provide valuable insights into the processes and habits regarding the production and use of images in the research context. © 2020, Springer Nature Switzerland AG.

Abstract
This paper develops a novel short-term load forecasting technique to predict the demanding power for the next hour. In this study, a linear equation-error Auto Regressive Auto Regressive Moving Average Exogenous (ARARMAX) model is trained to specify power consumption as a function of a few past hours. The parameters of the candidate mathematical model are estimated by using two least squares-based iterative algorithms. The main difference with these algorithms is the total number of past data involved in the modeling. Whereas practical data are always subject to noise and un-accurate measuring, a wavelet de-noising technique is utilized to reduce the effect of noise on forecasting which leads to more precise predictions. The superiority of the proposed approach is validated by utilizing practical data from a power utility in Canada in January 1995. The first three days' data are utilized to train the selected model and the fourth-day data are dedicated to test the prediction of the provided model. The L-2 and L-infinity norms error and MAPE, MAE, and RMSE are selected as criteria to show the merits of the proposed approach.

Abstract
The use of Educational Software (ES) in education has become essential for teachers and students. On the one hand, the effectiveness of its use may facilitate the acquisition of learning and on the other hand, it may enable a better transmission of the contents. In this sense, it is necessary to provide teachers with tools that allow them to develop successful pedagogical actions with appealing and innovative resources, capable of stimulating creativity and motivating students for learning. The aim of this study is to ascertain the knowledge and the use by teachers of ES Kahoot, GeoGebra and Texas Ti-Nspire, in what type of content, activities and what is the impact of their use in the teaching of Geometry and Measurement (GM), whether in teaching practice of teachers, or in the learning of students. The adopted method has a qualitative nature, with characteristics of a case study. Fourteen teachers who teach Mathematics at various schools in Portugal participated. Two questionnaires and a challenge that consisted of the elaboration of tasks were used as instruments. Data analysis was performed using Excel (Office 2016) and content analysis of the answers given, and the tasks developed. The results suggest that of the three ES, Kahoot was the most unknown and was the most chosen by teachers to develop different GM content. The reasons are also described as to why these ES may cause an improvement in the teaching practices of teachers, as well as motivation and student learning. © 2021, Springer Nature Switzerland AG.