Scholarlink

<p>Air pollution poses a significant threat to public health and environmental sustainability, particularly in urban areas with rapid industrialization and high vehicular emissions. This study focuses on Ghaziabad, one of India’s most polluted cities, to predict the Air Quality Index (AQI) using hourly environmental parameters such as pollutant concentrations, temperature, wind speed, and humidity, recorded from three monitoring stations. The primary objective is to evaluate AQI prediction models that can assist policymakers and urban planners in taking timely measures to improve air quality and public health. Several Machine Learning (ML) algorithms are used for AQI prediction, including Linear Regression (LR), Random Forest Regressor (RFR), Long Short-Term Memory (LSTM), and Extreme Gradient Boosting Regressor (XGBR). Among these, XGBR delivered the best performance, achieving a Mean Absolute Error of 2.536, a Mean Absolute Percentage Error of 1.62%, a Root Mean Square Error of 5.495, and an R<sup>2</sup> score of 0.997, indicating highly accurate predictions. Furthermore, these AQI values were classified into various categories using different models, in which the XGB classifier performed the best, achieving an accuracy of 97.62%. The findings demonstrate the efficacy of ML models in forecasting AQI with high precision using environmental variables. This research contributes directly to the United Nations Sustainable Development Goal (SDG) 11: Sustainable Cities and Communities by promoting data-driven urban air quality management. Integrating such predictive models into smart city frameworks is recommended to facilitate real-time air quality monitoring and proactive environmental policymaking.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Smart air quality assessment in Ghaziabad, India: machine learning approaches for AQI forecasting and classification

  • 期刊论文

  • 发表日期：

    2026年4月21日

查看全文
  • I. Dawar,
  • V. Singh,
  • T. Singhal,
  • M. Bhushan,
  • K. Agarwal

摘要

Air pollution poses a significant threat to public health and environmental sustainability, particularly in urban areas with rapid industrialization and high vehicular emissions. This study focuses on Ghaziabad, one of India’s most polluted cities, to predict the Air Quality Index (AQI) using hourly environmental parameters such as pollutant concentrations, temperature, wind speed, and humidity, recorded from three monitoring stations. The primary objective is to evaluate AQI prediction models that can assist policymakers and urban planners in taking timely measures to improve air quality and public health. Several Machine Learning (ML) algorithms are used for AQI prediction, including Linear Regression (LR), Random Forest Regressor (RFR), Long Short-Term Memory (LSTM), and Extreme Gradient Boosting Regressor (XGBR). Among these, XGBR delivered the best performance, achieving a Mean Absolute Error of 2.536, a Mean Absolute Percentage Error of 1.62%, a Root Mean Square Error of 5.495, and an R2 score of 0.997, indicating highly accurate predictions. Furthermore, these AQI values were classified into various categories using different models, in which the XGB classifier performed the best, achieving an accuracy of 97.62%. The findings demonstrate the efficacy of ML models in forecasting AQI with high precision using environmental variables. This research contributes directly to the United Nations Sustainable Development Goal (SDG) 11: Sustainable Cities and Communities by promoting data-driven urban air quality management. Integrating such predictive models into smart city frameworks is recommended to facilitate real-time air quality monitoring and proactive environmental policymaking.