Support Vector Machine Classification Algorithm Based on Relief-F Feature Weighting

Aiming at the problem that the existing support vector machines only consider the importance of samples and ignore the importance of features on the classification results, this paper proposes a support vector machine method based on Relief-F feature weighting. This method first uses the Relief-F feature weighting algorithm to calculate the weight value of each feature, and then uses the feature weight value to weight the inner product in the support vector machine kernel function. This method effectively avoids the influence of weakly correlated features or uncorrelated features on the support vector machine classification results. Training and verification are performed on the data provided by the UCI public data set; experimental results show that the method can improve the classification accuracy of the classifier, reduce the number of support vectors, and have better robustness and classification ability than traditional SVM.