Data extraction method, electronic equipment and media

Abstract

The present invention relates to a data selection method, electronic equipment and media, including obtaining period result data and all first data component information corresponding to the period result data, where the first data component information is continuous period random data; based on the Regression is performed on the cycle result data and all first data component information corresponding to the cycle result data to obtain the marginal goodness of fit corresponding to each of the first data component information, and the marginal goodness of fit corresponding to the first data component information Refers to the goodness-of-fit increment for the entire regression process after the first data component information is added to the regression; sort the marginal goodness-of-fit corresponding to all the first data component information in descending order, and select Information about the first M first data components, where M is a preset number. The present invention can accurately and quickly select the data component with explanatory degree to the periodic result data from massive continuous periodic random data components.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a data selection method, electronic equipment and media. Background technique [0002] The field of data processing is an important branch of the computer field. In the computer field, data can include text data, image data, audio data, video data, etc. according to presentation methods; according to storage methods, data can be stored in databases, text files, and specific format files (such as .doc / . xls), etc.; according to the way the data is formed, it can include static data and dynamic data, and dynamic data is especially data that changes with time, such as temperature data obtained through temperature sensors, network traffic data obtained through network devices such as routers, switches, etc. Device LBS data collected by GPS or Beidou, etc. Data processing can be "forward processing", such as using a certain algorithm to process the image to make it clearer, ...

AI Technical Summary

Problems solved by technology

There are also correlations between dynamic data, but because most of the dynamic data is data that changes over time (especially data that changes frequently over time), it is necessary to consider the impact of the time dimension on the correlation, which will not only occupy more computers Storage resources, retrieval resources, computing and processing resources, and the processing accuracy is not ideal, so how to efficiently obtain the relationship between dynamic data through "reverse processing" of data has become a difficult point in data processing

Images