From Literature Review to Extreme Weather Traffic Applications

Recently, I completed the construction of a Graph Neural Network (GNN) model applied to systematic literature review (SLR) tasks. This model integrates citation relationships and semantic similarity between publications to identify research topics and emerging trends.

Next, I will work on building a GNN model for traffic applications under extreme weather conditions. This project will:

• Construct a spatial-temporal graph representing road networks, traffic flow, and weather conditions.
• Integrate real-time traffic data with extreme weather information such as heavy rainfall, snowstorms, and high winds.
• Investigate how extreme weather disrupts traffic patterns, increases congestion risks, and impacts network resilience.

Recent Work: Literature Review GNN

The SLR-focused GNN has demonstrated strong potential for:

• Automating thematic clustering in large academic corpora.
• Providing interpretable outputs to support research synthesis.
• Highlighting emerging research areas and interdisciplinary connections.

Upcoming Work: Extreme Weather Traffic GNN

The next phase will adapt GNN techniques to model complex, dynamic traffic systems affected by severe weather.
Expected outcomes include:

• Improved travel time prediction under adverse conditions.
• Insights for urban planners and emergency response teams.

Stay Tuned

The transition from literature analysis to traffic resilience modeling will help explore the adaptability of GNN approaches across domains.