Predicting Air Quality based on Vehicle Congestion (CUSP 2024 Data Dive Best-Technical Winners)

Description

  • Juanes Lamilla

  • 23 February, 2023

Part of the 2024 Centre for Urban Science and Progress (CUSP) London Data Dive.

Predicted high precision street-level air quality through the use of congestion monitors by employing machine learning and spatial analysis techniques.

Awarded 'Best Technical Contribution' award, competing against teams from KCL, UCL, NYU, and Glasgow.

Utilized innovative use of models including Partial Least Squares (PLS), Seasonal Autoregressive Integrated Moving Average with Exogenous Inputs (SARIMAX), Long Short-Term Memory (LSTM) networks, and spatial analysis techniques like Moran's I and Geographically Weighted Regression (GWR).

Technology