Social networks’ role as primary information sources has led to a rise in misinformation, threatening the information ecosystem. This study introduces a system that assigns misinformation scores to tweets in real-time using a mix of Temporal Graph Network (TGN) and Recurrent Neural Networks (RNNs) to track how misinformation spreads. This approach uses active learning, a dual model system, a temporal embargo based on belief scores, and a retraining method, showing significant accuracy in identifying and reducing misinformation propagation across five datasets.

This paper introduces a comprehensive knowledge extraction and management framework aimed at tackling COVID-19 vaccination misinformation. The framework employs advanced natural language processing techniques, such as sarcasm detection and clustering, to analyze and categorize misinformation from online sources, requiring significantly fewer labeled tweets compared to existing methods. It effectively categorizes vaccine dissenting tweets, enhances discourse analysis, and outperforms baseline models in classification tasks.

The growth of social media underscores the need for early misinformation detection, as traditional methods often lag due to a lack of initial dissemination data. This paper introduces a model that classifies misinformation based on propagation paths and linguistic cues, and uses a causal user attribute inference to pinpoint potential misinformation spreaders or believers. The proposed model can accurately detect fake news within 30 minutes of its first appearance, achieving 86.5% accuracy before 50 retweets, surpassing current benchmarks.

We present a demonstration of nighttime light pattern (NTL) analysis system. Our tool named NightVIEW is powered by an efficient system architecture to easily export and analyse huge volume of spatial data (NTL), image segmentation and clustering algorithms to find unusual NTL patterns and identify hotspots of excess night light usage as well as finding semantics of cities.

This study explores the complex relationship between elephants and their environment, emphasizing its significance for ecology and conservation strategies. The research focuses on understanding elephant movement in Sub-Saharan Africa, considering ecological factors like seasons and rainfall patterns. Despite the challenges, the study offers valuable insights into elephant migratory behavior within the dynamic African landscape. By predicting how ecological determinants impact elephant migration, the research aids in informed conservation strategies, especially considering climate change effects on the environment. The findings contribute to minimizing human-elephant conflict, managing land use, and improving anti-poaching efforts for sustainable coexistence in the region.