User value scoring method and system based on level analysis algorithm

Abstract

The invention discloses a user value scoring method and system based on a level analysis algorithm, which relates to the field of user value scoring in live broadcast. The method comprises the following steps: establishing a level analysis structure of user values according to behavior information of users; performing the level analysis on each layer, calculating a weighing factor of each level-I index to the comprehensive value of each user, and calculating a weighing factor of each level-II index corresponding to the level-I index; and calculating a weight of all level-II indexes to the comprehensive value of each user according to the calculated weighing factors; calculating a score of the comprehensive value of each user by adopting a weight summing way; and finally selecting a corresponding score interval according to the calculated comprehensive value score of each user, and classifying a value level of each user according to the different score intervals. By adopting the method and the system, the automatic scoring of the user comprehensive values can be realized; and moreover, the scoring process is accurate and efficient, and the human cost is reduced.

Description

technical field [0001] The invention relates to the field of user value scoring in live video broadcasting, and specifically relates to a user value scoring method and system based on an analytic hierarchy process algorithm. Background technique [0002] With the rapid development of the live broadcast industry and the enrichment of platform services, the scale of users has shown a trend of rapid growth. Users have different behaviors on the platform, which bring different values ​​to the website, such as gift giving, recharge information, etc. How to evaluate the comprehensive value score of users, which are high-value users and which are low-value users, is convenient for marketers According to the different value levels of users, planning different marketing plans and retention and retention strategies is an inevitable problem to be solved. [0003] At present, when major websites evaluate the comprehensive value of current users, they still use manual screening of high-...

AI Technical Summary

Problems solved by technology

[0003] At present, when major websites evaluate the comprehensive value of current users, they still use manual screening of high-value and low-value users based on empirical rules, and there is no suitable algorithm for evaluating the comprehensive value of users.

Moreover, in actual operation, the method of manually screening high-value and low-value users is often carried out with a large degree of subjectivity, making the evaluation standards inconsistent; in addition, in the scenario of massive data, user behavior data There are often many dimensions and a large amount of data. Manually judging user classification standards is often inaccurate, the coverage rate is not high enough, repetitive work is easy to cause mistakes, and the manual operation time is long, the classification efficiency is low, and the labor cost is high.

Images