Abstract: In view of the complex state-related factors and fluctuations in the copper removal process,a status evaluation method based on feature-weighted fuzzy support vector machine(FSVM) is proposed. The fuzzy membership is used to define different weights for each sample,which can distinguish outliers from representative samples. The sample feature weights are calculated by information gain,improving the FSVM kernel function. The fuzzy SVM based on feature weights is then constructed for the status evaluation of copper removal. Using standard data sets and industrial data,the classification accuracies of different methods are compared,demonstrating the validity and accuracy of the proposed method.