An improved random forest method for air quality classification

Abstract

The invention discloses an improved random forest method for air quality classification, which comprises an air quality classification model, the air quality classification model comprises an originaldata module, a data preprocessing module, a classification generation module and a classification data module; the original data module is used for collecting original air data; the data preprocessing module is used for performing data cleaning, data integration, data conversion and other operations on the original air data; the classification generation module is used for randomly sampling the data processed by the data preprocessing module and classifying a decision tree based on a CART algorithm; the classification data module is used for receiving the classification model output by the classification generation module and outputting an air data classification result, and the random sampling comprises a self-service sampling method based on sample category grouping and a random featuresubspace method. According to the method, the classification precision of minority class samples is improved, and the overall misclassification cost of a sample set is reduced.

Description

technical field [0001] The invention relates to the technical field of air quality monitoring, in particular to an improved random forest method for air quality classification. Background technique [0002] Air pollution refers to a phenomenon in which human beings carry out production and life in the natural environment, and some pollutants are discharged into the atmosphere due to some inappropriate behaviors. When the concentration of substances reaches a certain value, it will cause harm to human health and the natural environment. The atmospheric environment is complex and changeable and has dynamic uncertain characteristics. There are many related factors that cause air pollution, such as PM(2.5), SO 2 and O 3 Such as air pollutants exceeding a certain concentration, as well as factors such as precipitation, wind direction, and humidity, all of which have a strong nonlinear relationship with the air quality in the future. More accurate air quality forecasts can help ...

AI Technical Summary

Problems solved by technology

When it classifies unbalanced data, due to the imbalance in the number of samples of the majority class and the samples of the minority class, the classification model will be biased towards the majority class and ignore the minority class, resulting in low classification accuracy of the minority class, which will cause training problems. The error development of the data set eventually leads to wrong data classification; it affects the indirect judgment of users who use sample data, and then misleads people's production and life, which has a high cost of misclassification

Images