Extracting mineralized information based on ACO_SVM

Abstract

Tuning the h y perparameters of Support Vector Machine (SVM) is an important wa y to improve the generalization performance of SVM. Because it is time consuming to seek for optimal parameters b y grid search method, the Ant Colon y Optimization (ACO) search method selecting parameters is presented in m y paper, which can acquire the best parameters of SVM .. M y research was done in Xijieri area, Qinghai province. Firstl y , anal y zed the behavior of SVM is anal y zed that have influence on the classrate; Secondl y , ACO algorithm is used for is designed and implemented,which is used for seeking for optimal SVM parameters; In the end, mineralized information is extracted b y ACO _ SVM. The experimental results show that ACO algorithm can seek for parameters quicklier and more satisfactoril y than grid search method; Through on the spot investigation and comparing with data of the known alteration areas in the mineral geolog y mapping of field, we find that extracting mineralized information b y ACO _ SVM is a good wa y .

Study on Preparation of Water Treatment Ceramic Particles by Using Sludge and Fly Ash and the Ceramic Particles’ Application

“…Along with the urbanization and social economic development, urban sewage's quantity continues to increase, resulting in increasing emission of life sludge [1]. The sludge contains massive organic pollutant, heavy metal, pathogen, viruses and so on, and thus has a strong polluting nature.…”