Method for detecting abnormal value of financial data based on kernel PCA and KMeans algorithm

The invention discloses a financial data abnormal value detection method based on a kernel PCA and a KMeans algorithm, and the method comprises the following steps: 1, carrying out the dimension reduction of high-dimensional financial data through employing a principal component analysis kernel PCA,...

Full description

Saved in:
Bibliographic Details
Main Authors LIU MINGFENG, LI JIANKE, LIANG LEI, HU YAN, CHEN ZAIDIE
Format Patent
LanguageChinese
English
Published 23.02.2024
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The invention discloses a financial data abnormal value detection method based on a kernel PCA and a KMeans algorithm, and the method comprises the following steps: 1, carrying out the dimension reduction of high-dimensional financial data through employing a principal component analysis kernel PCA, mapping original data to a high-dimensional space, projecting the first K principal components, and carrying out the projection of the first K principal components, obtaining data after dimension reduction; step 2, carrying out normalization processing on the data after dimension reduction; and step 3, selecting the number of clusters and specifying a distance threshold value from data to the center of mass of the clusters during abnormal value detection. According to the method, the kernel PCA and the KMeans algorithm are combined to carry out data abnormal value detection, dimension reduction is carried out on the high-dimensional financial data through the kernel PCA, and then the KMeans algorithm is utilized t
Bibliography:Application Number: CN202311547461