关键词: 连续小波变换方法, 主成分分析法, 最优化理论, 分类平面

Abstract: In order to improve the data utilization rate and avoid the influence of risk factors in information on intelligence analysis, a risk feature screening algorithm for intelligent intelligence analysis data based on SVM (Support Vector Machine) is proposed. The continuous wavelet transform method is used to eliminate the influence of noise signals in intelligence data on the analysis results, and the projection matrix is established by combining principal component analysis method to extract the main features of various types of noise-free intelligence data. The main feature extraction results of various kinds of intelligence data are input into support vector machine, and the classification plane in support vector machine is established by using optimization theory, and the classification rules of feature data in the classification plane are defined to screen the risk features of intelligence data. The experimental results show that the proposed method can accurately classify intelligence data, and the risk data detection efficiency is high, which can realize effective screening of risk data.

Key words: continuous wavelet transform method, principal component analysis method, optimization theory, classification plane