Data index screening method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and provides a data index screening method and device, computer equipment and a storage medium, and the method comprises the steps of carrying out the standardization processing of user data under a corresponding data index according to data distribution; generating a standard data vector of each user according to the standard data obtained through standardization processing, and extracting multiple pieces of index feature data in the standard data vector of each user; screening a plurality of pieces of first key index feature data according to the correlation indexes of the plurality of pieces of index feature data; extracting a plurality of pieces of second key index feature data and the index weight of each piece of second key index feature data by adopting a minimum absolute contraction and selection operator; and according to the index weight, using a Monte Carlo simulation method to simulate and train the user level prediction model for multiple times, and screening out multiple pieces of target key index feature data from the second key index feature data according to a prediction result corresponding to simulation training. According to the invention, the optimal data index can be screened out, and the screening efficiency of the data index is high.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a data index screening method, device, computer equipment and storage medium. Background technique [0002] In the face of a large number of data indicators, manual screening or traditional correlation analysis and screening methods can easily miss low-saturation but very important data indicators. The wrong choice of data indicators will lead to inaccurate effects of subsequent systems or models. . [0003] Especially when the data indicators are growing geometrically, manual screening or traditional correlation analysis is difficult to find all the potential correlations between a large number of data indicators, and the screened data indicators are mixed with more useless data Indicators, not only will not improve the effect of the model, but the number of selected data indicators is large, and it will easily lead to a very low efficiency of model construction...

AI Technical Summary

Problems solved by technology

[0002] In the face of a large number of data indicators, manual screening or traditional correlation analysis and screening methods can easily miss low-saturation but very important data indicators. The wrong choice of data indicators will lead to inaccurate effects of subsequent systems or models.

[0003] Especially when the data indicators are growing geometrically, manual screening or traditional correlation analysis is difficult to find all the potential correlations between a large number of data indicators, and the screened data indicators are mixed with more useless data Indicators, not only will not improve the effect of the model, but the number of selected data indicators is large, and it will easily lead to very low efficiency of model construction

Images