Behavioral Analysis and Sub-Topic Detection in COVID-19 Data Using Path-Aware Graph Neural Networks

The COVID-19 pandemic has significantly affected human behaviour and communication patterns, necessitating the development of advanced analytical techniques to decode these alterations. This work presents a unique way for performing thorough behavioural analysis and sub-topic detection on COVID-19 datasets using Path-Aware Graph Neural Networks (PAGNN). Our technique, which models complicated links and routes within the data, provides profound insights into how behaviours and debates evolved during the pandemic. We create extensive graphics from a variety of data sources, including social media interactions, transportation records, and health data, to provide a comprehensive picture of the pandemic's impact. The PAGNN model, combined with path-aware mechanisms, excels at discovering subtle patterns and theme structures that standard techniques frequently ignore. Our experimental results suggest that the PAGNN model outperforms conventional strategies in detecting nuanced behavioural shifts and emergent sub-topics. These findings provide significant views for public health authorities and policymakers, allowing them to make more informed decisions in the face of current and future health crises. This work emphasizes the revolutionary potential of advanced graph neural network designs in public health analytics, offering a powerful framework for deriving actionable insights from large, multidimensional datasets.