To address the Grounwater contamination, CrowdWaterSens, an uncertainty-aware graph neural network framework, examines the uncertainty and spatial irregularity of the crowdsensing groundwater contamination data and its relevant anthropogenic context to accurately estimate groundwater nitrate concentration.

Results from this study contribute to define a complete set of environmental and social data and information, which can help European decision makers to define new criteria for sustainable management of the waste plastics of interest. A new methodological approach has been proposed: it appears able to be applied in future research projects involving innovative management options.

This paper explores the potential implementation of the Consumption Footprint rationale to define a footprint indicator for the EU Bioeconomy, henceforth ‘Bioeconomy Footprint’. This indicator can be a powerful tool for a comprehensive and effective monitoring of the bioeconomy sectors: to capture environmental impacts over time, identifying environmental hotspots, highlighting geographic and sectorial trade-offs, and identifying burden shifts among impact categories and along the supply chain.

This study aims to identify the factors that constrain and enable the sustainability of reusable packaging systems, considering environmental, economic, social and technical dimensions. This research is critical to the effective implementation and scale-up of reusable packaging systems.

The main idea of this work is: To provide a brief review of India's capability for the deployment of carbon capture, utilization, and storage technology on a large scale.

This article is an in-depth analysis of the challenges and solutions for the circular economy. This article relates to SDG 12, Responsible Consumption and production.

This study investigates three Mediterranean coastal lagoons to study harmful algae and pathogens on plastic debris.

The article introduces a novel approach called PTM-SFFS that effectively combines pre-trained transformer models (PTMs) with traditional statistical classifiers and linguistic features for the task of gender screening on community question-answering (cQA) platforms.

In the context of applying machine learning to solve problems for risk prediction, disease detection, and treatment evaluation, EHR pose many challenges– they do not have a consistent, standardized format across institutions particularly in US, can contain human errors and introduce collection biases. In addition, some institutions or geographic regions do not have access to the technology or financial resources necessary to implement EHR, thus resulting in vulnerable and disadvantaged communities not being electronically visible.