Unlock instant, AI-driven research and patent intelligence for your innovation.

A Siamese Autoencoder Neural Network Algorithm and System for Accelerated Recommendation

A neural network algorithm and autoencoder technology, which is applied in the twin autoencoder neural network algorithm and system field, can solve the problems of limited recommendation speed, large disk memory space occupation, huge number of users and products, etc., to achieve accurate recommendation, Small space, fast effect

Active Publication Date: 2022-06-07
CHONGQING UNIV
View PDF15 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a twin autoencoder neural network algorithm and system for accelerated recommendation, which aims to solve the problem of using real values ​​to represent users and products occupying a large number of disks due to the huge number of users and products in the prior art. Memory space, and the technical problem of real value-based recommendation speed limitation

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
  • A Siamese Autoencoder Neural Network Algorithm and System for Accelerated Recommendation
  • A Siamese Autoencoder Neural Network Algorithm and System for Accelerated Recommendation
  • A Siamese Autoencoder Neural Network Algorithm and System for Accelerated Recommendation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] see Figure 1 to Figure 3 , the present invention provides a twin autoencoder neural network algorithm for accelerated recommendation, comprising the following steps:

[0027] S1: Obtain the basic recommendation system, and map users and products to the low-dimensional latent factor space;

[0028] S2: Pre-train the twin autoencoder neural network algorithm to obtain low-dimensional binary representations of users and products for discrete recommendations;

[0029] S3: Integrate the encoder into the existing recommendation system to improve the recommendation speed.

[0030] Among them, in step S1, the basic recommendation system is obtained. Take the collaborative filtering recommendation system based on matrix decomposition as an example. Its core is to map users and products to the low-dimensional latent factor space, and then the user's preference for products can use both The inner product of indicates that high similarity between user and item factors will lead ...

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 present invention relates to the field of artificial intelligence technology, in particular to a twin autoencoder neural network algorithm and system for accelerated recommendation; the neural network architecture takes twin neural networks as the main body, and two weight-sharing units are autoencoders, thereby A neural network is implemented to jointly learn user and item representations. The present invention is friendly to existing recommendation systems and does not need to retrain the entire model. When the recommendation system processes recommendation tasks based on similarity search, it is faster and requires less space than common recommendation methods; The discrete recommendation method recommends more accurately.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a twin autoencoder neural network algorithm and system for accelerating recommendation. Background technique [0002] With the development of computer science and technology, the types and quantities of online goods and services have increased dramatically, and it has become increasingly difficult for users to find interesting goods. The recommendation system saves the time and effort for users to find products by pre-matching users and products, and improves the efficiency of online product service platforms. It is recognized as an effective method to solve the problem of user information overload, and is used by a large number of online shopping malls, audio and video websites. Collaborative filtering based on matrix decomposition is an important method in the recommendation system. It decomposes the rating matrix of users on products into a low-dimensional hid...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/04G06N3/08G06Q30/06
CPCG06N3/084G06Q30/0631G06N3/045
Inventor 危枫陈蜀宇胡飞李军杨文武赵进
Owner CHONGQING UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com