Detecting Elevated Air Pollution Levels by Monitoring Web Search Queries: Algorithm Development and Validation

• Published in: JMIR formative research Vol. 6; no. 12; p. e23422
• Main Authors: Lin, Chen, Yousefi, Safoora, Kahoro, Elvis, Karisani, Payam, Liang, Donghai, Sarnat, Jeremy, Agichtein, Eugene
• Format: Journal Article
• Language: English
• Published: Canada JMIR Publications 19.12.2022
• Subjects: Air pollution; Epidemiology; Infectious diseases; Original Paper; Outdoor air quality; United States > US; web-based public health surveillance; search engine log analysis; neural network sequence modeling; nowcasting of air pollution; mobile phone; air pollution exposure assessment
• ISSN: 2561-326X; 2561-326X
• DOI: 10.2196/23422

Real-time air pollution monitoring is a valuable tool for public health and environmental surveillance. In recent years, there has been a dramatic increase in air pollution forecasting and monitoring research using artificial neural networks. Most prior work relied on modeling pollutant concentrations collected from ground-based monitors and meteorological data for long-term forecasting of outdoor ozone (O ), oxides of nitrogen, and fine particulate matter (PM ). Given that traditional, highly sophisticated air quality monitors are expensive and not universally available, these models cannot adequately serve those not living near pollutant monitoring sites. Furthermore, because prior models were built based on physical measurement data collected from sensors, they may not be suitable for predicting the public health effects of pollution exposure. This study aimed to develop and validate models to nowcast the observed pollution levels using web search data, which are publicly available in near real time from major search engines. We developed novel machine learning-based models using both traditional supervised classification methods and state-of-the-art deep learning methods to detect elevated air pollution levels at the US city level by using generally available meteorological data and aggregate web-based search volume data derived from Google Trends. We validated the performance of these methods by predicting 3 critical air pollutants (O , nitrogen dioxide, and PM ) across 10 major US metropolitan statistical areas in 2017 and 2018. We also explore different variations of the long short-term memory model and propose a novel search term dictionary learner-long short-term memory model to learn sequential patterns across multiple search terms for prediction. The top-performing model was a deep neural sequence model long short-term memory, using meteorological and web search data, and reached an accuracy of 0.82 (F -score 0.51) for O 0.74 (F -score 0.41) for nitrogen dioxide, and 0.85 (F -score 0.27) for PM , when used for detecting elevated pollution levels. Compared with using only meteorological data, the proposed method achieved superior accuracy by incorporating web search data. The results show that incorporating web search data with meteorological data improves the nowcasting performance for all 3 pollutants and suggest promising novel applications for tracking global physical phenomena using web search data.