Exploring Air Quality in Harbin and Zhengzhou by Linear Regression Approach

In recent years, straw burning has become a major source of air pollution, especially in cities like Harbin and Zhengzhou, where seasonal burning leads to spikes in PM2.5 levels. This study looks into how linear regression can be used to assess air quality, focusing on its ability to measure relationships between different factors and predict pollution levels. By collecting real-world pollution data and factoring in weather conditions, this research applies multiple linear regression to explore how different variables interact and improve prediction accuracy through statistical validation. The results show that this method works better than traditional approaches, offering more reliable estimates. In Zhengzhou, time-weighted regression performed even better, with an R of 97.78%, proving its strength in tracking pollution trends over time. This approach not only makes predictions more precise but also provides solid data to guide environmental policies and pollution control strategies. The study highlights linear regression as a useful tool for air quality analysis and lays the groundwork for future machine learning applications.