A Study of Principal Component Analysis in Comprehensive Indicator Screening for Parocneria Orienta Hazard

The aim of this paper is to investigate the application of principal component analysis in impact factors of Parocneria orienta hazard in Zhongjiang County. A principal component analysis was performed by the data of 12 factors associating with the prediction of P. Orienta hazard. The degree of P. Orienta hazard in different regions was obtained and intuitively presented by GIS. The eigenvalues of 3 principal components were 35.274%、20.544%、14.897%, respectively. The accumulative contribution rate of the 3 major factors to total variation accounted for 70.715%, maintaining most of information of 12 characters. According to the comprehensive value of principal component, GIS Natural Breakpoint Method was used to divide the degree of P. Orienta hazard into frequently occurring region, occasionally occurring region and safe area. Results showed that principal component analysis could optimize the comprehensive indicators for the evaluation of P. Orienta hazard, and comprehensive score of Principal Component could quantify and intuitively show the degree of P. Orienta hazard in different regions.