4G user loss early warning method and system based on multi-dimensional combined variables

Abstract

The invention provides a 4G user loss early warning method and system based on multi-dimensional combined variables, and the method comprises the steps of obtaining a significant index which influences user loss, wherein the significant index comprises a consumption transaction index, a competitor influence index, an abnormal silence index, a remote roaming index, and a use time index; and calculating a loss probability through the trained user loss early warning model, and if the loss probability is greater than a set threshold, determining that the user is a to-be-lost user, and concluding into five influence dimensions, namely a consumption transaction index, a competitor influence index, an abnormal silence index, a remote roaming index and a use time index. A principal component analysis idea is applied to recombine and integrate a plurality of variables in a dimension into a modeling index with the maximum information value, so that the problems of collinearity of an estimation equation and loss of original data information are solved, the value of a service index is maximized, and the data prediction accuracy is improved.

Description

Technical field [0001] The present invention relates to the field of communication technology, and more specifically, to a 4G user loss early warning method and system based on multi-dimensional combination variables. Background technique [0002] The trend of full-service operation has intensified market competition among communication operators. In order to gain more market share, operators are trying their best to reduce user churn rate and reduce the occurrence of user disconnection. Research shows that if a company’s user churn rate is reduced by 5%, profits will increase by 25% to 85%. The cost of acquiring a new user for a company is relatively high, but sometimes it only requires a phone call or a strategy to retain a user. . For this reason, operators need to accurately locate user attributes and try to retain user resources that tend to be lost. 4G network technology has brought about the explosive growth of data. 4G user resources have become the core resources of op...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a 4G user churn early warning method and system based on multi-dimensional combination variables to overcome the above problems or at least partly solve the above problems, which solves the problem of lag in user churn and retention decision-making, insufficient processing efficiency, and the prediction process in the prior art. Problems such as collinearity of explanatory variables and loss of original variable data information

Images