Back

A Graph Embedding Approach for Deciphering the Longitudinal Associations of Global Mobility and COVID-19 Cases

Awasthi, R.; Modi, M.; Dudeja, H.; Bajaj, T.; Rastogi, S.; Sethi, T.

2023-04-03 health informatics
10.1101/2023.03.30.23287969 medRxiv
Show abstract

The COVID-19 pandemic has highlighted the importance of monitoring mobility patterns and their impact on disease spread. This paper presents a methodology for developing effective pandemic surveillance systems by extracting scalable graph features from mobility networks. We utilized Metas "Travel Patterns" dataset to capture the daily number of individuals traveling between countries from March 2020 to April 2022. We have used an optimized node2vec algorithm to extract scalable features from the mobility networks. Our analysis revealed that movement embeddings accurately represented the movement patterns of countries, with geographically proximate countries exhibiting similar movement patterns. The temporal association dynamics between Global mobility and COVID-19 cases highlighted the significance of high-page rank centrality countries in mobility networks as a key intervention target in controlling infection spread. Our proposed methodology provides a useful approach for tracking the trajectory of infectious diseases and developing evidence-based interventions.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.