Search

Energy communities can support the energy transition, by engaging citizens through collective energy actions and generate positive economic, social and environmental outcomes. Renewable Energy Sources (RES) are gaining increasing share in the electricity mix as the economy decarbonises, with Photovoltaic (PV) plants to becoming more efficient and affordable. By incorporating Artificial Intelligence (AI) techniques, innovative applications can be developed to provide added value to energy communities. In this context, the scope of this paper is to compare Machine Learning (ML) and Deep Learning (DL) algorithms for the prediction of short-term production in a solar plant under an energy cooperative operation. Three different cases are considered, based on the data used as inputs for forecasting purposes. Lagged inputs are used to assess the historical data needed, and the algorithms' accuracy is tested for the next hour's PV production forecast. The comparative analysis between the proposed algorithms demonstrates the most accurate algorithm in each case, depending on the available data. For the highest performing algorithm, its performance accuracy in further forecasting horizons (3 hours, 6 hours and 24 hours) is also tested. [ABSTRACT FROM AUTHOR]

Animal Vehicle Collision (AVC) is relatively an evolving source of fatality resulting in the deficit of wildlife conservancy along with carnage. It's a globally distressing and disturbing experience that causes monetary damage, injury, and human-animal mortality. Roadkill has always been atop the research domain and serendipitously provided heterogeneous solutions for collision mitigation and prevention. Despite the abundant solution availability, this research throws a new spotlight on wildlife-vehicle collision mitigation using highly efficient artificial intelligence during nighttime hours. This study focuses mainly on arousal mechanisms of the "Histogram of Oriented Gradients (HOG)" intelligent system with extracted thermography image features, which are then processed by a trained, convolutional neural network (1D-CNN). The above computer vision – deep learning-based alert system has an accuracy between 94%, and 96% on the arousal mechanisms with the empowered real-time data set utilization. [ABSTRACT FROM AUTHOR]

Reliable and rapid non-invasive testing has become essential for COVID-19 diagnosis and tracking statistics. Recent studies motivate the use of modern machine learning (ML) and deep learning (DL) tools that utilize features of coughing sounds for COVID-19 diagnosis. In this paper, we describe system designs that we developed for COVID-19 cough detection with the long-term objective of embedding them in a testing device. More specifically, we use log-mel spectrogram features extracted from the coughing audio signal and design a series of customized deep learning algorithms to develop fast and automated diagnosis tools for COVID-19 detection. We first explore the use of a deep neural network with fully connected layers. Additionally, we investigate prospects of efficient implementation by examining the impact on the detection performance by pruning the fully connected neural network based on the Lottery Ticket Hypothesis (LTH) optimization process. In general, pruned neural networks have been shown to provide similar performance gains to that of unpruned networks with reduced computational complexity in a variety of signal processing applications. Finally, we investigate the use of convolutional neural network architectures and in particular the VGG-13 architecture which we tune specifically for this application. Our results show that a unique ensembling of the VGG-13 architecture trained using a combination of binary cross entropy and focal losses with data augmentation significantly outperforms the fully connected networks and other recently proposed baselines on the DiCOVA 2021 COVID-19 cough audio dataset. Our customized VGG-13 model achieves an average validation AUROC of 82.23% and a test AUROC of 78.3% at a sensitivity of 80.49%. [ABSTRACT FROM AUTHOR]

B Information b is widely available and accessible, but frequently leads to information overload and overexposure, while the effort for coding, storing, hiding, securing, transmitting and retrieving it may be excessive. Special Collection of Extended Selected Papers on "Novel Research Results Presented in The 12th International Conference on Information, Intelligence, Systems and Applications (IISA2021), 12-14 July 2021, Chania, Crete, Greecehttps://easyconferences.eu/iisa2021/" The authors employ machine learning-based methodologies to formulate a general evaluation chart of nutritional biomarkers. [Extracted from the article]

Item Response Theory (IRT) evaluates, on the same scale, examinees who take different tests. It requires the linkage of examinees' ability scores as estimated from different tests. However, the IRT linkage techniques assume independently random sampling of examinees' abilities from a standard normal distribution. Because of this assumption, the linkage not only requires much labor to design, but it also has no guarantee of optimality. To resolve that shortcoming, this study proposes a novel IRT based on deep learning, Deep-IRT, which requires no assumption of randomly sampled examinees' abilities from a distribution. Experiment results demonstrate that Deep-IRT estimates examinees' abilities more accurately than the traditional IRT does. Moreover, Deep-IRT can express actual examinees' ability distributions flexibly, not merely following the standard normal distribution assumed for traditional IRT. Furthermore, the results show that Deep-IRT more accurately predicts examinee responses to unknown items from the examinee's own past response histories than IRT does. [ABSTRACT FROM AUTHOR]