Multivariable data classification method and device

Abstract

The invention discloses a multivariable data classification method and device, and the method comprises the steps: carrying out the preprocessing of original data, and performing data filling according to a K-proximity algorithm and generate data to be extracted using the to-be-extracted data to train a data classification model classified according to the data features; and further performing data classification on the data by using a data classification model. According to the technical scheme, different original data or different types of original data can be processed and classified, andeffective information extraction of multivariable complex data is achieved.

Description

technical field [0001] The present invention relates to the field of computers, and more specifically, relates to a multivariate data classification method and device. Background technique [0002] With the development of computer technology and artificial intelligence technology, deep learning technology has more processing methods and application cases in organized data such as images, audio, and text, such as traditional RNN (recurrent neural network) and CNN (convolutional neural network). Network), etc., can achieve coarse-grained classification for some data with relatively high discrimination. However, as deep learning technology gradually enters a variety of industries, the complexity of data processing doubles, and data is often unorganized, unrelated between different dimensions, or there are a large number of missing and abnormal data. The random discarding mechanism cannot effectively eliminate abnormal data, nor can it realize intelligent identification and sim...

AI Technical Summary

Problems solved by technology

However, as deep learning technology gradually enters a variety of industries, the complexity of data processing doubles, and data is often unorganized, unrelated between different dimensions, or there are a large number of missing and abnormal data. The random discarding mechanism cannot effectively eliminate abnormal data, nor can it realize intelligent identification and similar restoration of missing data, which leads to poor fitting of classification models and makes it difficult to identify the classification of data and extract hidden information from data.

In short, the existing technology is still unable to effectively handle such complex data scenarios

[0003] There is currently no effective solution to the problem of difficult data classification in complex data scenarios in existing technologies

Images