SVM based sub-classification study of engineering waste glass

With the promotion of urbanization in China, a large amount of construction waste materials are piled up. In order to solve the problem of idle construction waste materials, We takes construction waste glass as an example, and establishes a SVM multi-classification prediction model based on one-to-many from two perspectives of classification prediction accuracy and model sensitivity. The data of chemical composition content in waste glass is divided into training set test set according to the ratio of 8:2 to complete the training and validation of the model. The data were brought into the trained model, resulting in five categories. Finally the model was analysed for sensitivity by Monte Carlo simulation with an accuracy of 95.4%.