Scholarlink

Today, the governments of developing countries are focusing more on air pollution. Air pollution occurs because of factors like fuel use in vehicles such as private transport, industries, and, more importantly, the burning of waste like plastic and grass. These are a few of the elements that contribute to air pollution. All of these factors have an impact on air quality. It can be calculated or determined by using PM2.5 and other variables. If this level is high, then it may be a serious problem for people’s health. To control air pollution, continuous pollution monitoring is required. Here, different machine learning models like linear regression, decision tree classifiers, random forest classifiers, and KNN classifiers are used to detect whether the sample of air is polluted or not. On the basis of previous readings of PM2.5, NH3, CO, NO, NOx, and NO2, SO2 is used to predict the future values of the AQI. By collecting datasets of the daily atmospheric conditions of a specific city or town, this system is able to predict the PM2.5 level and also detect the air pollution status.

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

Machine Learning-Based Air Pollution Monitoring and Forecasting

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Naga Ravindra Babu M,
  • M. Durga Satish,
  • B. V. Prasanthi,
  • S. V. V. D. Jagadeesh,
  • J. N. S. S. Janardhana Naidu,
  • Immidi Kali Pradeep

摘要

Today, the governments of developing countries are focusing more on air pollution. Air pollution occurs because of factors like fuel use in vehicles such as private transport, industries, and, more importantly, the burning of waste like plastic and grass. These are a few of the elements that contribute to air pollution. All of these factors have an impact on air quality. It can be calculated or determined by using PM2.5 and other variables. If this level is high, then it may be a serious problem for people’s health. To control air pollution, continuous pollution monitoring is required. Here, different machine learning models like linear regression, decision tree classifiers, random forest classifiers, and KNN classifiers are used to detect whether the sample of air is polluted or not. On the basis of previous readings of PM2.5, NH3, CO, NO, NOx, and NO2, SO2 is used to predict the future values of the AQI. By collecting datasets of the daily atmospheric conditions of a specific city or town, this system is able to predict the PM2.5 level and also detect the air pollution status.