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

Domain-adaptive non-overlapping entity cross-domain data collaborative item recommendation algorithm

A data collaboration and recommendation algorithm technology, which is applied in the fields of electronic digital data processing, digital data information retrieval, special data processing applications, etc., can solve problems such as data distribution differences, knowledge transfer inconsistencies, etc., to improve recommendation performance, ensure consistency, and high The effect of predicting performance

Pending Publication Date: 2022-07-05
ZHEJIANG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems existing in the background, the present invention provides a cross-domain collaborative recommendation method based on domain-adaptive non-overlapping entities for the current cross-domain recommendation method that does not consider the inconsistency of knowledge transfer caused by the difference in data distribution between domains

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
  • Domain-adaptive non-overlapping entity cross-domain data collaborative item recommendation algorithm
  • Domain-adaptive non-overlapping entity cross-domain data collaborative item recommendation algorithm
  • Domain-adaptive non-overlapping entity cross-domain data collaborative item recommendation algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0118] Attached below figure 1 The present invention will be further described in detail with specific examples.

[0119] like figure 1 As shown, the embodiment and process thereof implemented according to the complete method of the content of the present invention are as follows:

[0120] 1) Select related auxiliary field data according to the target field data. For example, if the target field data is the movie field, the book field or the music field can be selected as the auxiliary field data.

[0121] 2) By preprocessing the user behavior data in the target domain data and auxiliary domain data, the target domain data preference matrix R is obtained t and the auxiliary domain data preference matrix R a . target preference matrix R t It is randomly divided into training set and test set according to a certain proportion.

[0122] 3) Calculate the similarity matrix between data users in the target domain and the similarity matrix between items and the similarity m...

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 domain-adaptive cross-domain data collaborative item recommendation algorithm without overlapping entities. Preprocessing the target domain data and the auxiliary domain data to obtain respective user-article preference matrixes and user-article indication matrixes; processing to obtain a similarity matrix between the users and between the articles; constructing a user graph and an article graph according to the similarity matrix; establishing a target function with a graph regularization item, and solving the target function to obtain three optimal parameter matrixes; and reconstructing by using a product of the three parameter matrixes to obtain a user-article recommendation matrix, and recommending articles to the user by using the user-article recommendation matrix. According to the method, data distribution among the fields is aligned by using a field adaptive technology, so that the consistency of knowledge migration is ensured, and the recommendation performance in the target field data can be improved.

Description

technical field [0001] The invention belongs to a cross-domain data recommendation method in the field of cross-domain recommendation algorithms, and in particular relates to a domain-adaptive cross-domain data collaborative item recommendation algorithm without overlapping entities. Background technique [0002] With the rapid development of information technology, e-commerce sites (such as Amazon, JD.com, Taobao, etc.), online video sites (such as Tencent Video, iQiyi, Youku, etc.), music sites (such as NetEase Cloud Music and QQ Music, etc.) and Content on news sites (such as Toutiao and Tencent News) has grown exponentially, and the need for efficient, real-time recommendations has grown stronger. Personalized recommendation systems aim to provide users with personalized online product or service recommendations to solve the increasingly serious problem of information overload. Collaborative filtering technology is one of the most popular methods in recommendation syste...

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): G06F16/9536G06F16/9535G06V10/74G06K9/62
CPCG06F16/9536G06F16/9535G06F18/22
Inventor 孔祥维张洪为
Owner ZHEJIANG UNIV
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