Genetic Projection Pursuit Interpolation Data Mining Model for Urban Environmental Quality Assessment
Computer Science and Information Engineering, World Congress on (2009)
Los Angeles, California USA
Mar. 31, 2009 to Apr. 2, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.6
In order to solve the incompatibility problem of assessment data indexes and to raise the precision of assessment model for urban environmental quality, a genetic projection pursuit interpolation data mining model (GPPIDMM) is presented for comprehensive assessment of urban environmental quality. In this model the projection pursuit data mining, genetic algorithm, interpolation curve and the assessment standards of urban environmental quality are used. And the indexes values of urban environmental quality can be synthesized to one dimension projection values. The samples can be assessed according to the values of the projection values in one dimension space. 50, 100, 500, 1000 samples in each grade have been adopted to test the stability of parameters in this model. In this new model, 5000 samples are generated from the assessment standards of urban environmental quality, which avoids the low precision in other models with little quantity of samples. The interpolation assessment formula is given with projection values and experiential grades. And the parameters in this model are steady by test. This new model is used to assess Xuanzhou environmental quality with the main indexes of water environment, atmospheric environment and noise environment. GPPIDMM can also be used to design the weights of the index system and deal with data. The results show that the urban environmental quality is still clean in Xuanzhou. GPPIDMM is a new method for evaluation of urban environmental quality and it is more objective in the whole data processing.
W. Wei, S. Dunxian, Y. Xiaohua, L. Jianqiang and H. Xiaoxue, "Genetic Projection Pursuit Interpolation Data Mining Model for Urban Environmental Quality Assessment," 2009 WRI World Congress on Computer Science and Information Engineering, CSIE(CSIE), Los Angeles, CA, 2009, pp. 805-809.