Order processing method, device and medium

Abstract

The disclosure provides an order processing method, device and medium. The method includes: acquiring order data of items satisfying preset conditions within a specific period, each order data corresponding to an order of an item; acquiring user advertising behavior data, the user advertising behavior data including order user advertisement Behavioral data and user advertising behavior data without orders; use the user advertising behavior data as the input of the neural network, and use the corresponding value of the actual generated value of the order corresponding to each data in the user advertising behavior data as a preset answer, train The neural network is used to predict whether an order will be generated, and the training is completed when the consistency between the prediction of the neural network and the preset answer reaches a predetermined standard; and the trained neural network is output. The method also includes using the trained neural network to predict the contribution rate of various advertisements to orders and total sales of items.

Description

technical field [0001] The present disclosure relates to the field of Internet technology, and more specifically, to an order processing method, device and medium. Background technique [0002] With the development of the Internet, the types of advertisements (including distribution channels, display methods, etc.) are becoming more and more diverse. The user's final purchase behavior may be influenced by one or several types of advertisements, and different types of advertisements may have different degrees of influence. In the prior art, in order to analyze how various types of advertisements affect the user's final purchase behavior and the degree of influence, a shallow model (such as a Markov model) or a rule model (such as a custom model, such as based on time decay, average Distribution principles, etc.) for analysis. [0003] In the process of implementing the concept of the present invention, the inventors found that there are at least the following problems in th...

AI Technical Summary

Problems solved by technology

[0003] In the process of implementing the concept of the present invention, the inventors found that there are at least the following problems in the prior art: the shallow model is weak in extracting deep correlation features of the data between various types of advertisements and the user's final purchase behavior, and the expressive ability Limited; the rule model cannot mine the inherent characteristics of the data between various types of advertisements and the user's final purchase behavior, so it cannot provide deeper data association information

It can be seen that the analysis of the association relationship between various types of advertisements and the user's final purchase behavior in the prior art is not accurate and in-depth.

Images