Content recommendation model training method, content recommendation method and related device

Abstract

The invention discloses a content recommendation model training method, a content recommendation method and a related device, which are used for improving the diversity of recommended content. The method comprises the steps of obtaining first to-be-trained text data at a first moment, the first to-be-trained text data comprising a first text attribute feature and a first text semantic feature; obtaining a first text quality prediction value and a first feature vector corresponding to the first to-be-trained text data through a neural network of the to-be-trained content recommendation model; obtaining a second text quality prediction value and a second feature vector corresponding to second to-be-trained text data at a second moment through the neural network; obtaining a text recommendation probability through a determinant point process according to the first text quality prediction value, the first feature vector, the second text quality prediction value and the second feature vector; and according to the text recommendation probability, training the to-be-trained content recommendation model to obtain a trained content recommendation model.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a method for training a content recommendation model, a method for content recommendation and related devices. Background technique [0002] With the rapid development of the Internet, people's daily life is increasingly inseparable from the Internet. Listening to music, watching videos, and browsing information through the Internet have become people's daily habits. Taking browsing information content as an example, with the explosive growth of various information content (such as news, articles and video content, etc.) Actively recommending interesting information content has become a feasible and efficient solution. [0003] At present, the content push method is mainly to tag the content, and then make relevant recommendations based on the number of user clicks. However, if this recommendation method is used for a long time, the entire recommended page will have ...

AI Technical Summary

Problems solved by technology

[0003] At present, the content push method is mainly to tag the content, and then make relevant recommendations based on the number of user clicks. However, if this recommendation method is used for a long time, the entire recommended page will have the same type of content, and the content will be diverse. poor

Images