Data processing method and device, electronic equipment and storage medium

Abstract

The invention provides a data processing method and device, electronic equipment and a storage medium. The method comprises: obtaining high-dimensional data, the number of to-be-selected features, a first target weight matrix and a second target weight matrix, wherein the high-dimensional data comprises a plurality of features; selecting a first feature subset from the high-dimensional data, wherein the first feature subset comprises a number of features to be selected; determining the score of each feature in the high-dimensional data according to the high-dimensional data, the first feature subset, a first target weight matrix and the second target weight matrix; according to the score of each feature in the high-dimensional data, selecting a second feature subset from the high-dimensional data, the second feature subset comprising a number of features to be selected; and determining the second feature subset as a target feature subset. According to the method, the data processing accuracy can be improved, computer resources are saved, and the data processing efficiency is improved.

Description

technical field [0001] The present disclosure relates to the field of computer technology, and in particular to a data processing method, device, electronic equipment and storage medium. Background technique [0002] When classifying high-dimensional data, because there are a large number of irrelevant and redundant features in high-number data, it will lead to very high computational costs and reduce the performance of tasks such as classification accuracy or clustering accuracy. [0003] In related technologies, calculating the score of each feature subset in a brute force calculation method usually requires a high computational cost; using a single high-dimensional data as a vector, that is, by continuously concatenating all its row or column vectors, the high-dimensional If the dimension data is converted into a long vector, it will fall into the disaster of dimensionality because the dimension of the sample matrix will be much larger than the number of samples, which wi...

AI Technical Summary

Problems solved by technology

[0002] When classifying high-dimensional data, because there are a large number of irrelevant and redundant features in high-number data, it will lead to very high computational costs and reduce the performance of tasks such as classification accuracy or clustering accuracy.

[0003] In related technologies, calculating the score of each feature subset in a brute force calculation method usually requires a high computational cost; using a single hi

Images