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

A Personalized Recommendation Method Based on Recurrent Generative Adversarial Networks

A recommendation method and network technology, applied in the computer field, can solve problems in the field of recommendation that have not yet been applied, and achieve the effect of avoiding the artificial feature extraction process

Active Publication Date: 2021-06-01
HANGZHOU NORMAL UNIVERSITY
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Generative confrontation networks have been used a lot in the field of image generation, but they have not been applied to the field of recommendation, especially in movie recommendation

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 Personalized Recommendation Method Based on Recurrent Generative Adversarial Networks
  • A Personalized Recommendation Method Based on Recurrent Generative Adversarial Networks
  • A Personalized Recommendation Method Based on Recurrent Generative Adversarial Networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055] A specific implementation case of the present invention is further described below in conjunction with the accompanying drawings, as figure 1 Shown:

[0056] The data used in this example is MovieLens, which was founded by the GroupLens project team of the School of Computer Science and Engineering at the University of Minnesta in the United States. This dataset describes unlimited tags of movies within 5 stars for user recommendations. The dataset contains 100,004 ratings for 9,066 movies by 671 users. Users are randomly selected, and each selected user rates at least 20 movies.

[0057] Step (1). Construct matrix image:

[0058] Sampling the time series of the user's historical behavior with the method of equidistant sliding window and constructing matrix image sample set R:

[0059]

[0060] t is the total number of samples, n is the number of users, m is the number of items;

[0061] Taking 30 days as a cycle, a two-dimensional matrix image is generated ever...

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 relates to a personalized recommendation method based on cyclic generation confrontation network. The method of the present invention constructs a matrix sample incorporating time information by sampling historical user behavior based on time information, and uses a cyclically generated confrontation network to automatically extract features and display dynamic timing behavior. The method of the present invention first constructs a matrix image, constructs a cyclically generated confrontation network, then cyclically generates the internal process of the confrontation network, trains the cyclic generation of the confrontation network, and finally sorts the row pixels of a certain user in the figure, and sorts the first N columns with large values is displayed as the recommended list for this user. The method of the present invention can transmit time information from the cycle of generating information, slice time, and each model is an independent network connected by information transmission, which can better reflect time information in terms of model principles and practical effects, Reactive time series patterns.

Description

technical field [0001] The invention belongs to the field of computer technology, and relates to a personalized recommendation method based on a cyclic generative confrontation network to predict the user's preference degree for item items. Background technique [0002] Personalization is one of the areas receiving the most attention right now. Today, with the rapid development of science and technology, more and more products make the market oversupply, and consumers need to choose their favorite products from many similar products. But now people pay more attention to the internal quality and price of products, and pay more attention to the diversification of material product forms, internalization and popularization of quality. With the increase of people's income, the shortening of popular consumption period and product life cycle, the imitation speed of fashionable consumption methods is accelerated, and the consumption concept is gradually developing towards the exter...

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): G06F16/9535G06Q30/06
CPCG06Q30/0631
Inventor 张子柯王睿修李丽霞刘闯
Owner HANGZHOU NORMAL UNIVERSITY
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