Scholarlink

This research focuses on predicting and analyzing long-term air quality trends of various cities in India, whose data are combined into a single dataset, using deep learning time series models such as LSTM, Bi-LSTM, and statistical time series models such as ARIMA and SARIMA. Further, we provide a comparative analysis of their performance. In this study, we focus on particulate matter (PM \(_{2.5}\) and PM \(_{10}\) ), evaluating the capacity of each model to predict pollution levels effectively. It aims to capture both short-term fluctuations and seasonal patterns in air quality. Analysis of data for a period from 2019 to 2023, covering both pre- and post-COVID periods, highlighted a notable decrease in particulate matter during the pandemic session. This reduction is likely to be due to lower industrial output, fewer vehicles on the roads and limited human activities at the time of lockdowns.

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

Time Series Deep Learning Models to Predict and Analyze Air Quality Index of Indian Cities

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Devarpita Dey,
  • Kuldip Katiyar

摘要

This research focuses on predicting and analyzing long-term air quality trends of various cities in India, whose data are combined into a single dataset, using deep learning time series models such as LSTM, Bi-LSTM, and statistical time series models such as ARIMA and SARIMA. Further, we provide a comparative analysis of their performance. In this study, we focus on particulate matter (PM \(_{2.5}\) and PM \(_{10}\) ), evaluating the capacity of each model to predict pollution levels effectively. It aims to capture both short-term fluctuations and seasonal patterns in air quality. Analysis of data for a period from 2019 to 2023, covering both pre- and post-COVID periods, highlighted a notable decrease in particulate matter during the pandemic session. This reduction is likely to be due to lower industrial output, fewer vehicles on the roads and limited human activities at the time of lockdowns.