Research On Urban Air Quality And Urban Development In China Based On Data Mining

The basic theory of data mining clustering method and classification method is mainly introduced,and the compare between the two is made.Based on the clustering and classification methods,the clustering analysis of the examples is carried out by using the shortest distance method,the longest distance method,the Ward method and the class mean method in the system clustering method,and the clustering results obtained by different methods are compared.The recursive segmentation tree and the C5.0 algorithm in the classification method are used to classify and analyze the examples,and a reasonable classification decision tree is obtained according to the actual situation.Firstly,the air quality data of 30 major cities(except Lasa)in China are taken as the research object.Seven annual pollutants are used,namely,the average annual concentration of sulfur dioxide,the average annual concentration of nitrogen dioxide,the average annual concentration of respirable particulate matter,The annual average concentration of fine particles,soot emissions,carbon monoxide daily average 95 percentile concentration and ozone day maximum 8 hours 90 percentile concentration,through the data mining method and statistical software R language,the clustering model is eatablished,the 30 cities is divided into two categories,the good air quality cities,including Haikou,Kunming,Nanchang,Nanning,Guiyang,Fuzhou and Lanzhou,other cities' air quality are poor.Then,according to the clustering results,eight urban development indicators related to air quality(motor vehicle volume,vehicle volume per unit area,million motor vehicle ownership,added value of secondary industry,industrial added value above scale,per capita consumption Expenditure,per capita GDP,greening rate)using the data mining method and statistical software R language to do decision tree classification analysis,extracted the urban development of urban air quality is more significant indicators,obtained decision tree classification model specific rules.In particular,the main factors influencing the air quality in urban development can be obtained from the decision tree.The greening rate of pollutants,pollutants and pollutants in cities and municipalities above the scale,and the impact of human factors on urban air is not significant.Take a further step we can also get the city's area has become a major problem in air pollution control.Finally,according to the empirical analysis of this paper,the corresponding suggestions are put forward,is to set the area of large cities together for jointly control and multi-regional operations;improve the car energy structure and promote public transport can help to reduce motor vehicle pollution;by focusing on high-tech industries,Class of ordinary industries,can help to reduce the proportion of heavy industry to improve the industrial structure;combined with the characteristics of each city area to develop a different green city policy guidance,improve and supervise the evaluation system.