Short video click rate prediction method based on sequence capsule network

Abstract

The invention discloses a short video click rate prediction method based on a sequence capsule network. According to the method, based on a click sequence of a user to a short video, multiple interests of the user are mined by utilizing a sequence capsule network, and the click rate of the user to a target short video is predicted. The method is mainly composed of three parts: in the first part, context features are extracted from a user click sequence by using a convolutional neural network; in the second part, a sequence capsule network is used for converting the context features into different interest spaces, the seriousness of user behaviors is captured in the different interest spaces, and multi-interest vector representation of the user is obtained; and in the third part, the clickrate of the short video is predicted based on the multi-interest vector representation of the user.

Description

technical field [0001] The invention belongs to the technical field of Internet services, and in particular relates to a method for predicting the click-through rate of short videos based on sequence capsule networks. Background technique [0002] Short video is a new type of short video. The shooting of short videos does not require the use of professional equipment or professional skills. Users can easily shoot and upload to the short video platform directly through their mobile phones, so the number of short videos on the short video platform is growing very fast. This makes the demand for an effective short video recommendation system very urgent. An effective short video recommendation system can improve user experience and user stickiness, thereby bringing huge commercial value to the platform. [0003] In recent years, many researchers have proposed video-based personalized recommendation methods. These methods can be classified into three categories: collaborative...

AI Technical Summary

Problems solved by technology

[0004] The above methods all take into account the sequence of the user's short video click sequence, but they do not take into account the diverse interests of the user

When the above methods model user interests, they directly treat user interests as a whole, which will lose a lot of information, especially the user's main interest will cover up the user's secondary interest

Images