Abstract

Abstract
This paper presents BATS-PT, the manual translation of the lexicographic portion of the Bigger Analogy Test Set (BATS) to European Portuguese. BATS-PT covers ten types of lexico-semantic analogies and can be used for assessing word embeddings and language models. Following this, the dataset is showcased while assessing two pretrained language models for Portuguese, BERTimbau and Albertina, in two tasks: analogy solving and relation completion, both in zero- and few-shot mask-prediction approaches. Experiments reveal different performance across relations and, in both tasks, the best overall performance was achieved with BERTimbau, in a five-shot scenario. We further discuss the limitations of the reported experiments and directions towards future improvements in these tasks. © 2024 The Association for Computational Linguistics.

Abstract
In the modern world, making healthy food choices is increasingly important due to the rise in food-related illnesses. Existing tools, such as Nutri-Score and comprehensive food labels, often pose challenges for many consumers. This paper proposes an application that uses Optical Character Recognition (OCR) technologies to read and interpret food labels, thus upgrading current solutions that rely mainly on reading product barcodes. By using advanced optical character recognition and machine learning techniques, the system aims to accurately extract and analyze nutritional information directly from food packaging without relying on a database of pre-registered products. This innovative approach not only increases consumer awareness, but also supports personalized diet management for diseases such as diabetes and hypertension, while promoting healthier eating habits and better health outcomes. Two minimalist functional prototypes were developed as a result of this work: a desktop application and a mobile application.

Abstract
Model checkers can automatically verify a system’s behavior against temporal logic properties. However, analyzing the counterexamples produced in case of failure is still a manual process that requires both technical and domain knowledge. However, this step is crucial to understand the flaws of the system being verified. This paper presents a language created to support the generation of natural language explanations of counterexamples produced by a model checker. The language supports querying the properties and counterexamples to generate the explanations. The paper explains the language components and how they can be used to produce explanations.

Abstract
Since the 1970s, extensive halite extraction in Macei & oacute;, Brazil, has resulted in significant geological risks, including ground collapses, sinkholes, and infrastructure damage. These risks became particularly evident in 2018, following an earthquake, which prompted the cessation of mining activities in 2019. This study investigates subsidence deformation resulting from these mining operations, focusing on the collapse of Mine 18 on 10 December 2023. We utilized the Quasi-Persistent Scatterer Interferometric Synthetic Aperture Radar (QPS-InSAR) technique to analyze a dataset of 145 Sentinel-1A images acquired between June 2019 and April 2024. Our approach enabled the analysis of cumulative displacement, the loss of amplitude stability, the evolution of amplitude time series, and the amplitude change matrix of targets near Mine 18. The study introduces an innovative QPS-InSAR approach that integrates phase and amplitude information using amplitude time series to assess the lifecycle of radar scattering targets throughout the monitoring period. This method allows for effective change detection following sudden events, enabling the identification of affected areas. Our findings indicate a maximum cumulative displacement of -1750 mm, with significant amplitude changes detected between late November and early December 2023, coinciding with the mine collapse. This research provides a comprehensive assessment of deformation trends and ground stability in the affected mining areas, providing valuable insights for future monitoring and risk mitigation efforts.

Abstract
Wireless capsule endoscopy (WCE) has revolutionized the field of gastrointestinal examinations, being MedtronicTM WCE one of the most used in clinics. In those WCE videos, medical experts use RAPID READERTM tool to annotate findings in videos. However, the frame annotations are not available in an open format and, when exported, they have different resolutions and some annotated artefacts that make difficult their localization in the original videos. This difficult the use of WCE medical experts' annotations in the research of new computed-aid diagnostic (CAD) methods. In this paper, we propose a methodology to compare image similarities and evaluate it in a private MedtronicTM WCE SB3 video dataset to automatically identify the annotated frames in the videos. We used state-of-the-art pre-trained convolutional neural network (CNN) models, including MobileNet, InceptionResNetv2, ResNet50v2, VGG19, VGG16, ResNet101v2, ResNet152v2, and DenseNet121, as frame features extractors and compared them with the Euclidean distance. We evaluated the methodology performance on a private dataset consisting of 100 WCE videos, totalling 905 frames. The experimental results showed promising performance. The MobileNet model achieved an accuracy of 94% for identifying the first match, while the top 5, top 10, and top 20 matches were identified with accuracies of 94%, 94%, and 98%, respectively. The VGG16 and ResNet50v2 models also demonstrated strong performance, achieving accuracies ranging from 88% to 93% for various match positions. These results highlight the effectiveness of our proposed methodology in localizing target frames and even identifying similar frames very use useful for training data-driven models in CAD research. The code utilized in this experiment is available on the Github† © 2024 The Author(s). Published by Elsevier B.V.