Research on the Application of Support Vector Machines in High-Dimensional Data Mining

This study aims to analyze the application of support vector machine (SVM) in high-dimensional data mining, analyze its specific practices in data preprocessing, feature selection, classification, regression and dimensionality reduction, and adopt the feature dimension of the Iris dataset and the kernel function of SVM to explore the performance of SVM in high-dimensional data classification tasks. First, standardization is used to preprocess the data, followed by polynomial feature expansion and cross-feature formation of high-dimensional data, and the radial basis function (RBF) kernel is used to map the data to a higher-dimensional feature space to further enhance the model effect. The experimental results show that after appropriate parameter optimization, SVM can significantly improve the classification accuracy, which can reach up to 96.5%. In addition, the experimental results also show that the penalty parameter C and kernel width $\sigma$ have a significant impact on the performance of the model. The conclusion is that SVM has strong robustness in the face of high-dimensional data, can effectively resolve the dilemma of high-dimensional data classification, and has shown wide application potential in a large number of application scenarios.