Abstract
Bird species identification is a relevant and time-consuming task for ornithologists and ecologists. With growing amounts of audio-annotated data, automatic bird classification using machine learning techniques is an important trend in the scientific community. Analyzing bird behavior and population trends helps detect other organisms in the environment and is an important problem in ecology. Bird populations react quickly to environmental changes, which make their real-time counting and tracking challenging and very useful. A reliable methodology that automatically identifies bird species from audio would therefore be a valuable tool for the experts in different scientific and applicational domains. The goal of this work is to propose a methodology to identify bird sounds. In this paper, we explore deep learning techniques that are being used in this domain, such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) to classify the data. In deep learning, audio problems are commonly approached by converting them into images using audio feature extraction techniques such as Mel Spectrograms and Mel Frequency Cepstral Coefficients (MFCCs). We propose and test multiple deep learning and feature extraction combinations in order to find the most suitable approach to this problem.

Abstract
Global models have been developed to tackle the challenge of forecasting sets of series that are related or share similarities, but they have not been developed for heterogeneous datasets. Various methods of partitioning by relatedness have been introduced to enhance the similarities of sets, resulting in improved forecasting accuracy but often at the cost of a reduced sample size, which could be harmful. To shed light on how the relatedness between series impacts the effectiveness of global models in real-world demand-forecasting problems, we perform an extensive empirical study using the M5 competition dataset. We examine cross-learning scenarios driven by the product hierarchy commonly employed in retail planning to allow global models to capture interdependencies across products and regions more effectively. Our findings show that global models outperform state-of-the-art local benchmarks by a considerable margin, indicating that they are not inherently more limited than local models and can handle unrelated time-series data effectively. The accuracy of data-partitioning approaches increases as the sizes of the data pools and the models' complexity decrease. However, there is a trade-off between data availability and data relatedness. Smaller data pools lead to increased similarity among time series, making it easier to capture cross-product and cross-region dependencies, but this comes at the cost of a reduced sample, which may not be beneficial. Finally, it is worth noting that the successful implementation of global models for heterogeneous datasets can significantly impact forecasting practice.

Abstract
The Phenobot platform is comprised by an autonomous robot, instrumentation, artificial intelligence, and digital twin diagnosis at the molecular level, marking the transition from pure data-driven to knowledge-driven agriculture 4.0, towards a physiology-based approach to precision viticulture. Such is achieved by measuring the plant metabolome 'in vivo' and 'in situ', using spectroscopy and artificial intelligence for quantifying metabolites, e.g.: i. grapes: chlorophylls a and b, pheophytins a and b, anthocyanins, carotenoids, malic and tartaric acids, glucose and fructose; ii. foliage: chlorophylls a and b, pheophytins a and b, anthocyanins, carotenoids, nitrogen, phosphorous, potassium, sugars, and leaf water potential; and iii. soil nutrients (NPK). The geo-referenced metabolic information of each plant (organs and tissues) is the basis of multi-scaled analysis: i. geo-referenced metabolic maps of vineyards at the macroscopic field level, and ii. genome-scale 'in-silico' digital twin model for inferential physiology (phenotype state) and omics diagnosis at the molecular and cellular levels (transcription, enzyme efficiency, and metabolic fluxes). Genome-scale 'in-silico' Vitis vinifera numerical network relationships and fluxes comprise the scientific knowledge about the plant's physiological response to external stimuli, being the comparable mechanisms between laboratory and field experimentation - providing a causal and interpretable relationship to a complex system subjected to external spurious interactions (e.g., soil, climate, and ecosystem) scrambling pure data-driven approaches. This new approach identifies the molecular and cellular targets for managing plant physiology under different stress conditions, enabling new sustainable agricultural practices and bridging agriculture with plant biotechnology, towards faster innovations (e.g. biostimulants, anti-microbial compounds/mechanisms, nutrition, and water management). Phenobot is a project under the Portuguese emblematic initiative in Agriculture 4.0, part of the Recovery and Resilience Plan (Ref. PRR: 190 Ref. 09/C05-i03/2021 - PRR-C05-i03-I-000134). © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/).

Abstract
Haptic feedback in Virtual Reality is commonly provided through wearable or grounded devices adapted to specific scenarios and situations. Shape-changing devices allow for the physical representation of different virtual objects but are still a minority, complex, and usually have long transformation times. We present Shape-a-getti, a novel ungrounded, non-wearable, and graspable haptic device that can quickly change between different radially symmetrical shapes. It uses a single actuator to rotate several identical poles distributed along a radius to render the different shapes. The format of the poles defines the possible shapes, and in our prototype, we used one that could render concave, straight, and convex shapes with different radii. We conducted a user evaluation with 21 participants asking them to recognize virtual objects by grasping the Shape-a-getti. Despite having difficulties distinguishing between some objects with very similar shapes, participants could successfully identify virtual objects with different shapes rendered by our device. (c) 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Abstract
The development of more flexible and robust models for reasoning about systems in environments with potentially conflicting information is becoming more and more relevant in different contexts. In this direction, we recently introduced paraconsistent transition systems, i.e. transition systems whose transitions are tagged with a pair of weights, one standing for the degree of evidence that the transition exists, another weighting its potential non existence. Moreover, these structures were endowed with a modal logic [3] that was further formalised as an institution in [5]. This paper goes a step further, proposing an approach for the structured specification of paraconsistent transition processes, i.e. paraconsistent transition systems with initial states. The proposed approach is developed along the lines of [12], which introduced a complete methodology for (standard) reactive systems development building on the Sannella and Tarlecki stepwise implementation process. For this, we enrich the logic with dynamic modalities and hybrid features, and provide a pallet of constructors and abstractors to support the development process of paraconsistent processes along the entire design cycle.

Abstract
Retailers depend on accurate forecasts of product sales at the Store x SKU level to efficiently manage their inventory. Consequently, there has been increasing interest in identifying more advanced statistical techniques that lead to accuracy improvements. However, the inclusion of multiple drivers affecting demand into commonly used ARIMA and ETS models is not straightforward, particularly when many explanatory variables are available. Moreover, regularization regression models that shrink the model's parameters allow for the inclusion of a lot of relevant information but do not intrinsically handle the dynamics of the demand. These problems have not been addressed by previous studies. Nevertheless, multiple simultaneous effects interacting are common in retailing. To be successful, any approach needs to be automatic, robust and efficiently scaleable. In this study, we design novel approaches to forecast retailer product sales taking into account the main drivers which affect SKU demand at store level. To address the variable selection challenge, the use of dimensionality reduction via principal components analysis (PCA) and shrinkage estimators was investigated. The empirical results, using a case study of supermarket sales in Portugal, show that both PCA and shrinkage are useful and result in gains in forecast accuracy in the order of 10% over benchmarks while offering insights on the impact of promotions. Focusing on the promotional periods, PCA-based models perform strongly, while shrinkage estimators over-shrink. For the non-promotional periods, shrinkage estimators significantly outperform the alternatives.