Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Mobile application recommendation method based on lightweight graph convolutional network

A mobile application and convolutional network technology, applied in the field of mobile applications, can solve problems such as increasing the difficulty of model training, reducing recommendation performance, over-smoothing effect, etc., and achieve the effect of accurate mobile application recommendation

Pending Publication Date: 2021-11-23
HUNAN UNIV OF SCI & TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are operations such as nonlinear activation and feature transformation in the existing mobile application recommendation based on graph convolutional neural network. These operations are used to model and represent the high-order interaction between users and mobile applications. The difficulty of model training leads to over-smoothing effect and reduces the recommendation performance

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Mobile application recommendation method based on lightweight graph convolutional network
  • Mobile application recommendation method based on lightweight graph convolutional network
  • Mobile application recommendation method based on lightweight graph convolutional network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The technical solution of the present invention will be described in further detail below in conjunction with specific implementation. The technical features or connection relationships described in the present invention are not described in detail. They are all existing technologies adopted.

[0033] Below in conjunction with embodiment, the present invention is described in further detail.

[0034] Such as Figure 1-3 As shown, the technical solution adopted by the present invention is as follows: a mobile application recommendation method based on a lightweight graph convolutional network, comprising the following steps:

[0035] 1) Initial embedding layer: embedding vector e i ∈R d represents the embedding matrix of user u, e i ∈R d Represents the embedding matrix of mobile application i, where d is the embedding dimension of the mobile application or user; because the embedding dimensions of the two are consistent, the parameter matrix is ​​constructed and inte...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a mobile application recommendation method based on a lightweight graph convolutional network; the method comprises the steps: firstly carrying out the modeling of an interaction relation between an APP and a user through a bipartite graph, and carrying out the initial embedding through the high-order connectivity (a tree diagram); then, in a lightweight graph convolutional layer, capturing a collaborative filtering signal along the graph structure using embedding propagation to further refine embedding of the mobile application with the user; and finally, predicting preferences of the user for different mobile applications through the inner product, and completing a recommendation task. The invention belongs to the technical field of mobile applications, and particularly relates to a mobile application recommendation method based on a lightweight graph convolutional network.

Description

technical field [0001] The invention belongs to the technical field of mobile applications, and specifically refers to a mobile application recommendation method based on a lightweight graph convolutional network. Background technique [0002] With the rapid growth of the number and types of mobile applications, how to accurately recommend mobile applications to users has become a new challenge. Graph Convolutional Neural Network is a typical recommendation technique for mobile applications. However, there are operations such as nonlinear activation and feature transformation in the existing mobile application recommendation based on graph convolutional neural network. These operations are used to model and represent the high-order interaction relationship between users and mobile applications. The difficulty of model training leads to over-smoothing effect and reduces the recommendation performance. [0003] Among them, lightweight graph convolutional network: LGC; Cont...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06N3/04G06K9/62G06F17/16G06F16/9535
CPCG06F17/16G06F16/9535G06N3/045G06F18/214
Inventor 曹步清钟为是
Owner HUNAN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com