An Automated Fine-Grained Air Quality Monitoring System Using Data Mining
Regardless of many years of progress, the air quality has begun to increase in the course of recent years. Air contamination has an enormous impact on people’s life despite several attempts to purify and reduce the impurities released. Our proposed system sets out an air quality forecasting model. The model involves the use of both BP neural network and data mining technique for individual purposes. Initially, this forecasting system uses the data mining technique to classify variables that have an effect on air quality. Additionally, this section utilizes these component’s information to set up a neural network. The BP neural network is used to analyze the impact factor and the data mining method is used to collect information on the factors that caused the depletion of air quality. Lastly, an evaluation of the forecasting model is executed. The test result shows that the air quality forecasting model built in this paper is now well constructed to resolve the issues, as it has a higher accuracy rate. Through analyzing and evaluating the results of the predictions, it is found that BP neural network algorithm has strong generalization capability and global search capability. This forecasting model enhances validity and feasibility and also provide accurate evidence in decision-making within the Department of Environmental Protection.
Keywords- air quality monitoring system, data mining, BP Neural Networks, forecasting model.