Decision classifier training method, audio and video recommendation method and device, and storage medium

Abstract

Embodiments of the invention relate to a decision classifier training method, an audio and video recommendation method and a device, and a storage medium. The method comprises the steps of obtaining historical operation data of a user for audios and videos from a log file; simultaneously carrying out model training on the plurality of different types of initial models based on the historical operation data to obtain a plurality of trained classification models; selecting a part of models from the plurality of classification models as candidate models according to a set rule; updating the weight of the historical operation data; training the candidate model by adopting the updated training sample; selecting a set number of target candidate models from the trained candidate models; generating a decision classifier according to a set number of target candidate models, the decision classifier integrates a plurality of target candidate models, compared with a single model, the prediction ofthe type of the user is more accurate, the accurate classification of the user can be realized by using the historical operation data of the user through the decision classifier, and subsequent recommendation of audios and videos for the user according to the type is facilitated.

Description

technical field [0001] The embodiments of the present invention relate to the field of audio and video processing, and in particular to a method, device and storage medium for training a decision classifier and recommending audio and video. Background technique [0002] With the development of network technology, film and television digital technology and hard disk storage technology, various network film and television systems are gradually emerging, and users usually use Internet film and television systems to watch audio and video. [0003] The Internet film and television system has a large library of film and television programs. Due to the massive library of film and television programs, most users will not and cannot spend a lot of energy to traverse all the film and television programs on the film and television system. Therefore, the film and television programs of the Internet film and television system Program recommendation is very important. [0004] In related...

AI Technical Summary

Problems solved by technology

[0004] In related technologies, the user's historical viewing records or user's historical search records are usually used to recommend movies and TV programs for users. This recommendation method cannot accurately predict the user's preferences and cannot achieve accurate recommendation of movies and TV programs.

Images