Abstract
The recent advances in natural language processing (NLP) are linked to training processes that require vast amounts of corpora. Access to this data is commonly not a trivial process due to resource dispersion and the need to maintain these infrastructures online and up-to-date. New developments in NLP are often compromised due to the scarcity of data or lack of a shared repository that works as an entry point to the community. This is especially true in low and mid-resource languages, such as Portuguese, which lack data and proper resource management infrastructures. In this work, we propose PT-Pump-Up, a set of tools that aim to reduce resource dispersion and improve the accessibility to Portuguese NLP resources. Our proposal is divided into four software components: a) a web platform to list the available resources; b) a client-side Python package to simplify the loading of Portuguese NLP resources; c) an administrative Python package to manage the platform and d) a public GitHub repository to foster future collaboration and contributions. © 2024 PROPOR. All Rights Reserved.

Abstract
Artificial intelligence (AI) has gained significant evolution in recent years that, if properly harnessed, may meet or exceed expectations in a wide range of application fields. However, because Machine Learning (ML) models have a black-box structure, end users frequently seek explanations for the predictions made by these learning models. Through tools, approaches, and algorithms, Explainable Artificial Intelligence (XAI) gives descriptions of black-box models to better understand the models' behaviour and underlying decision-making mechanisms. The AI development in companies enables them to participate in Industry 4.0. The need to inform users of transparent algorithms has given rise to the research field of XAI. This paper provides a brief overview and introduction to the subject of XAI while highlighting why this topic is generating more and more attention in many sectors, such as industry.

Abstract
The Czech Republic is home to the Hranice Abyss, the world's deepest natural underwater cave, a site extensively explored by a dedicated team of divers from a speleology group. Over the years, numerous studies have been conducted to unravel the cave's mysteries, delving into fields such as biology, hydrogeology, and geology. Mapping a cave of such vast dimensions and staggering depth poses formidable challenges, making the task hazardous, demanding, and timeintensive for a limited team of divers. In July 2022, the UNEXUP project was invited to explore and map the cave with its robot (UX1-neo), which contains many acoustic and optical sensors, used for navigation, localization, and mapping. Its unique control and dynamics allow the robot to successfully navigate through caves and flooded mines. This paper delves into the specifics of the six days of mission dives, offering insights into the mapping process, and presenting some of the results obtained from the entire cave.

Abstract
The adoption of sustainable dietary patterns that consider simultaneously nutritional well-being and reduced environmental impact is of paramount importance. This paper introduces the Dietary Pattern Sustainability Index (DIPASI), as a method to assess the sustainability of dietary patterns by covering the environmental, nutritional, and economic dimensions in a single score. Environmental indicators include carbon footprint, water footprint, and land use, the nutritional quality is evaluated through the Nutritional Rich Diet 9.3 score, and the economic aspects are considered using diet cost. DIPASI measures the deviation (in %) of an individual's diet in relation to a reference diet. The case study utilized dietary data from the Portuguese National Food, Nutrition, and Physical Activity Survey (IAN-AF 2015-2016), which included 2999 adults aged 18 to 64. The Portuguese dietary patterns (covering 1492 food products consumed), were compared against the reference Mediterranean diet. Results indicated that the Portuguese dietary pattern had a higher environmental impact (CF: 4.32 kg CO2eq/day, WF: 3162.88 L/day, LU: 7.03 m(2)/day), a lower nutritional quality (NRD9.3: 334), and a higher cost (6.65 euros/day) when compared to the Mediterranean diet (CF: 3.30 kg CO2eq/day, WF: 2758.84 L/day, LU: 3.67 m(2)/day, NRD9.3: 668, cost: 5.71 euros/day). DIPASI reveals that only 4% of the sample's population does not deviate or presents a positive deviation (> - 0.5%) from the Mediterranean diet, indicating that the majority of Portuguese individuals have lower sustainability performance. For the environmental sub-score, this percentage was 21.3%, for the nutritional sub-score was 10.9%, and for the economic sub-score was 34.2%. This study provides a robust framework for assessing dietary sustainability on a global scale. The comprehensive methodology offers an essential foundation for understanding and addressing challenges in promoting sustainable and healthy dietary choices worldwide.

Abstract
This scoping review systematizes the current research related to the use of both blockchain and machine learning techniques in medical imaging applications. A systematic electronic search was performed, and twenty-five studies were included in the review. These studies aimed to use blockchain and machine learning techniques to provide (i) efficient security mechanisms to support the communication of medical imaging data, (ii) aggregation of distributed medical imaging data to train machine learning algorithms, and (iii) machine learning algorithms based on federated learning strategies. Among the ten machine learning techniques identified in the included studies, Convolutional Neural Network was the most representative (i.e., 44% of the studies). Moreover, Artificial Neural Network, Capsule Network, Deep Neural Network, Gated Recurrent Units, and Neural Network were machine learning techniques used by more than one study. Although the included studies developed algorithms with potential impact in clinical practice, it must be noted that they did not discuss the generalizability of their algorithms in real-world clinical conditions.

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.