Article copywriting generation network training method, article copywriting generation method and device

Abstract

The embodiment of the invention discloses an article copywriting generation network training method, an article copywriting generation method and device, electronic equipment and a computer readable medium. A specific embodiment of the method comprises: obtaining article description information of each article in an article set; performing data preprocessing on an article description information set corresponding to the article set to obtain a processed article description information set; training the initial first article copywriting generation network to obtain a trained first article copywriting generation network; and training the initial second article copywriting generation network by using a knowledge distillation method to obtain a trained second article copywriting generation network. According to the embodiment, the trained first article copywriting generation network is used for guiding and training the initial second article copywriting generation network to generate the article copywriting, so that the trained second article copywriting generation network can accurately and effectively generate the article copywriting according to the title information of the article and the attribute information of the article.

Description

technical field [0001] The embodiments of the present disclosure relate to the field of computer technology, and specifically relate to a network training method for generating article text, a method for generating article text, a device, an electronic device, and a computer-readable medium. Background technique [0002] At present, the traditional technical application of item search and recommendation has been unable to better meet the growing needs of users. When browsing recommender systems, users often face the problem of information explosion. Users hope to quickly find the products they need through some excellent item copywriting to save costs. The usual way is to manually analyze the content information such as the title and attributes of the item and the comment information of the item to generate the item copy. [0003] However, when the above method is used to generate item copy, there are often the following technical problems: most items have less comment inf...

AI Technical Summary

Problems solved by technology

[0003] However, when the above method is used to generate item copy, there are often the following technical problems: most items have less comment information or the value of the comment information is low, and it is impossible to effectively generate relatively high-quality item copy based on the comment information

Images