Research Of Urban Air Quality Evaluation Based On Machine Learning

In recent years,city air pollution problems become more prominent, the direct discharge of automobile exhaust, industrial waste gas, dust and other large city to city environment, far more than the self purification capacity of the environment, leading to decline in air quality, direct crisis to the health and safety of city residents,the city air quality problem has attracted great attention from all walks of life Chinese the. In order to control the air pollution effectively,improve the city’s air quality evaluation, first of all must be reasonable on the air quality of city environment,science,and environmental protection departments of the city residents more objective understanding of city environment air quality, make a reasonable living arrangements and scientific prevention and control measures. Therefore, the evaluation of urban environmental air quality plays an important role in the prevention and control of air pollution.With the advent of big data era and the emergence of artificial intelligence, the traditional evaluation method cannot satisfy the needs of intelligent sensor data processing of massive demand, a large number of researchers based on data, to evaluate the city environment by using the intelligent method of air quality,machine learning is a branch of artificial intelligence. The machine learning is introduced to the city environmental air quality assessment,to evaluate the city environment using the random forest algorithm of machine learning in the air quality,through the training of random forest model, find the intrinsic mapping between various air pollutants and air quality level relations,establish a random forest evaluation model,improve the scientific evaluation and robustness.Based on the simulation results, the model with support vector machine, Naive Bayesian and K nearest neighbor models were compared, the simulation data of Shanghai city from 2013 to 2015 the air quality data, simulation results obtained good results, the experimental results show that the evaluation method of the best effect, the accuracy of up to 99.69%, at the same time the performance of the random forest model are analyzed to verify the applicability and stability of the evaluation method,it can be seen from the analysis of the characteristics of the generalization error of this model is not very sensitive to the number of variables, and the accuracy and time complexity has a good compromise between city environment, can be used for accurate and effective evaluation of the air quality.