Data processing method and related device

Abstract

The embodiment of the invention discloses a data processing method and a related device. The historical content push data of a target product having correlation with a to-be-promoted product is acquired for the to-be-promoted product, through the historical content push moment and the historical actual conversion moment in the historical content push data, information about how long the historical pushed object is converted after the content is pushed can be obtained, when a survival model is trained based on the object features of the historical pushed object, the data dimension of the conversion duration is introduced, the probability that the object is converted in a plurality of continuous time periods is concerned, the problem of conversion is converted into the problem of delayed conversion, learning of the survival model on delayed conversion in the time dimension is enhanced, decoupling of strong correlation between original historical content push data and a target product is achieved, and in a cold start stage, the first survival model obtained through training can temporarily replace a conversion rate model to provide prediction of the conversion rate for the product to be promoted.

Description

technical field [0001] The present application relates to the field of data processing, in particular to a data processing method and related devices. Background technique [0002] For a new product, it is necessary to increase the exposure of the product through the Internet through content push, etc., so that a large number of users can learn about the product through the pushed content in a short period of time, so as to quickly increase the audience of the product and promote the product. Conversion rate (or clickthrough rate, CVR) is a measure of the probability of successfully converting a user through content push. Successful conversion usually refers to the user's behavior of obtaining the product corresponding to the content. If the conversion rate corresponding to the user can be accurately predicted before the content is pushed, It can effectively improve the efficiency of content push, and increase the audience of new products as much as possible with limited pus...

AI Technical Summary

Problems solved by technology

However, this positive example may be converted after the new product is launched. However, in many cases, the actual reaction of users after the new product is launched is not as popular as when making an appointment. As a result, a large proportion of reserved users do not actually convert. Some of the positive examples for training the booking model are actually not positive examples, such as unconverted booking users

[0005] As a result, the reservation model based on reservation behavior training did not accurately learn the user's preference for new products. The CVR value predicted by the reservation model will be significantly higher than the actual situation. In the cold start stage, the reservation model will directly Reduce the effect of subsequent content push

Images