A recommendation method and system based on knowledge learning

A recommendation method and knowledge learning technology, applied in the field of user recommendation, can solve the problems of sparse rating matrix and inability to improve the recommendation effect, and achieve the effect of good recommendation results, good recommendation service experience, and improved recommendation effect.

Active Publication Date: 2021-02-09
CHENDU PINGUO TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional method of coordinated filtering to establish a scoring matrix cannot take a suitable way to effectively integrate other behaviors with target rows and item information to provide more information that is helpful for recommendation, so the scoring matrix of a single target behavior is often too sparse , cannot improve the recommendation effect

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 recommendation method and system based on knowledge learning
  • A recommendation method and system based on knowledge learning
  • A recommendation method and system based on knowledge learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0060] In this example, if Figure 1~3 As shown, a recommendation method based on knowledge learning, the method includes the following steps:

[0061] S1. Extract users and items as entities, and extract user operation behaviors and item attributes as relationships to obtain user-item data;

[0062] S2. Convert the user-item data into triples to obtain triple data including entities and relationships;

[0063] S3. Store the triplet data in RDF, use transE to learn knowledge representation, convert entities and relationships into vector representations, and obtain user-item knowledge graphs; the relationships include relationships between users and items and relationship between items;

[0064] The knowledge map of this embodiment is as follows image 3 As shown, there are 3 users A, B and C, items 1, 2, 3, 4, 5 and 6, and user operation behaviors: browsing, adding to shopping cart and purchasing. This embodiment is only to illustrate the method. The set knowledge map is r...

Embodiment 2

[0092] A recommendation system comprising:

[0093] 1. Triple generation module: used to extract users and items as entities, and extract user operation behaviors and item attributes as relationships, obtain user-item data, convert the user-item data into triples, and obtain entities including and relational triple data; the triple generation module includes:

[0094] 1.1. Entity extraction module: used to extract users and items as entities;

[0095] 1.2. Relationship extraction module: used to extract user operation behaviors and item attributes as relationships;

[0096] 2. Atlas storage module: used to store the triplet data in a database;

[0097] 3. Knowledge representation learning module: used for knowledge representation learning, converting entities and relationships into vector representations, that is, obtaining user-item knowledge graphs; this embodiment is transE;

[0098] 4. Recommendation module: Associate the vectors of entities and relationships in the use...

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 recommendation method based on knowledge learning, which belongs to the technical field of user recommendation. The method overcomes the shortcomings of the existing collaborative filtering technology. The recommendation method of the invention utilizes the interaction between more users and items integrated into the knowledge map data, and use the improved sub-graph embedding to improve the recommendation effect; the knowledge map provides a new idea for the fusion of heterogeneous data into the collaborative filtering algorithm to solve the problem of heterogeneity; the invention also provides a recommendation method for recommendation The system is convenient for users to quickly and timely generate recommendation lists, obtain better recommendation results, and provide users with a better recommendation service experience.

Description

technical field [0001] The invention relates to the technical field of user recommendation, in particular to a recommendation method and system based on knowledge learning. Background technique [0002] With the development of the Internet, various websites and apps have experienced information overload to varying degrees. How to choose suitable content from a lot of information to interested users is a problem faced by every website and app. The characteristics of the click message can be used to obtain the potential interest of the user, so as to push the message that the user is interested in to the user, thereby increasing the revenue of the website / App. [0003] Traditional personalized recommendation methods are based on user / item similarity, or based on collaborative filtering (matrix decomposition), and some hybrid methods. It may be considered that information is an item to be recommended. In the existing collaborative filtering technology, a scoring matrix of use...

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): G06F16/9535G06F16/36
Inventor 王丹徐滢
Owner CHENDU PINGUO TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap