Environment Data Analysis And Processing Based On Support Vector Machine

Abstract:In this thesis,based on the statistical principles,to Changsha, Zhuzhou and Xiangtan region’s soil as the background,it uses the support vector machine algorithm to analysis the relationship between the heavy metal elements and PH value in the soil,establish a mathematical model that it can predict the change of the content of heavy metals in the soil and design the model’s solution algorithm,based on matlab programming solution, according to the calculation results to analyze the relationship between the heavy metal content and pH value in the soil and compare with the results by the regression analysis model and BP neural network model. The main work and innovation of this article are as follows:(1)The thesis briefly introduce the background of this subject-the state of soil environment and the severe situation of the soil pollution in the modern, review support vector machine’s basic theory, models, algorithms and knowledge.(2)Take advantage of the multiple regression analysis and BP neural network method to analyzes the present soil situation of Changsha,Zhuzhou and Xiangtan area.In the soil there is a certain linear relationship with heavy metal content and the PH value from the multiple regression analysis, and BP neural network provides a predictive model.(3)Based on the support vector machine theory,the thesis researches and analyse on the samples,and establishes mathematical model. Research and design the corresponding algorithm programming to solve the model to predict the trend of heavy metals, and compare with the predict’s results via multiple regression analysis, BP neural network found that the results error is the smallest by the support vector machine,the generalization performance of the model is better,the complexity of the algorithm is smaller.(4)Summarize the full text, this thesis points out some shortcomings in the theory and applications and the further research direction of support vector machine.