Comparative Analysis of Machine Learning-Based Imputation Techniques for Air Quality Datasets with High Missing Data Rates

TL;DR

The paper tackles the challenge of predicting PM2.5 under spatiotemporal high missingness by comparing conventional ML, deep learning, and diffusion models on a Dublin, Ireland dataset collected from mobile and fixed sensors. It introduces a 500 m grid-based processing pipeline, augments data with external features like traffic and weather, and uses SMOTE to address severe class imbalance, achieving a notable 82% missing rate in the merged data. The study finds that diffusion models achieve the highest F1 score (0.9486) while ensemble methods, particularly random forest, provide the best overall accuracy (up to 94.82%), with external features enhancing several models’ performance. The work demonstrates practical pathways for robust air quality imputation and prediction in data-sparse urban environments and motivates future extensions to other pollutants, regions, and driver-support applications for vulnerable road users.

Abstract

Urban pollution poses serious health risks, particularly in relation to traffic-related air pollution, which remains a major concern in many cities. Vehicle emissions contribute to respiratory and cardiovascular issues, especially for vulnerable and exposed road users like pedestrians and cyclists. Therefore, accurate air quality monitoring with high spatial resolution is vital for good urban environmental management. This study aims to provide insights for processing spatiotemporal datasets with high missing data rates. In this study, the challenge of high missing data rates is a result of the limited data available and the fine granularity required for precise classification of PM2.5 levels. The data used for analysis and imputation were collected from both mobile sensors and fixed stations by Dynamic Parcel Distribution, the Environmental Protection Agency, and Google in Dublin, Ireland, where the missing data rate was approximately 82.42%, making accurate Particulate Matter 2.5 level predictions particularly difficult. Various imputation and prediction approaches were evaluated and compared, including ensemble methods, deep learning models, and diffusion models. External features such as traffic flow, weather conditions, and data from the nearest stations were incorporated to enhance model performance. The results indicate that diffusion methods with external features achieved the highest F1 score, reaching 0.9486 (Accuracy: 94.26%, Precision: 94.42%, Recall: 94.82%), with ensemble models achieving the highest accuracy of 94.82%, illustrating that good performance can be obtained despite a high missing data rate.

Figures (3)

• Figure 1: Comparison of hourly average pm2.5 values in 3 datasets, with time series decomposition of dpd data as an example.
• Figure 2: Spatial distribution of pm2.5 data in three datasets.
• Figure 3: Confusion matrix and roc curves for rf and ldm.