Supercharge Your Innovation With Domain-Expert AI Agents!

Similar information retrieval method and system based on isoproton map neural network

A neural network and information retrieval technology, applied in the field of similar information retrieval methods and systems based on heterogeneous subgraph neural networks, can solve the problems of reducing robustness, increasing costs, and difficult to improve the effect, so as to improve applicability, improve The effect of similarity

Active Publication Date: 2022-03-11
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF12 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Method 1 introduces additional information such as interaction records to improve the effect but increases the cost and reduces robustness, and is not suitable for low-resource scenarios; Method 2 does not have a targeted design model for similar content retrieval, and the effect is difficult promote

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
  • Similar information retrieval method and system based on isoproton map neural network
  • Similar information retrieval method and system based on isoproton map neural network
  • Similar information retrieval method and system based on isoproton map neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0101] The invention discloses a similar information retrieval method based on a heterogeneous subgraph neural network. figure 1 It is a flowchart of a similar information retrieval method based on a heterogeneous subgraph neural network according to an embodiment of the present invention, such as figure 1 As shown, the method includes:

[0102] Step S1. Extract the entities directly related to the business, use the entities as nodes, and construct the edges between nodes according to the semantic relationship between entities to complete the graph structured data; after completing the graph structured data, initialize an Embedding for each node , which initializes an embedding representation;

[0103] In some embodiments, in the step S1, the specific method of extracting entities directly related to the business, using the entities as nodes, and constructing edges between nodes according to the semantic relationship between entities includes:

[0104] use Represents a col...

Embodiment 2

[0190] The invention discloses a similar information retrieval system based on heterogeneous subgraph neural network. Figure 5 It is a structural diagram of a similar information retrieval system based on a heterogeneous subgraph neural network according to an embodiment of the present invention; as Figure 5 As shown, the system 100 includes:

[0191] The graph structured data module 101 is configured to extract entities directly related to the business, use the entities as nodes, and construct edges between nodes according to the semantic relationship between entities to complete the graph structured data;

[0192] The heterogeneous subgraph neural network model 102 is configured to include: a general subgraph paradigm neighborhood information modeling module, an information aggregation module for heterogeneous nodes, an information aggregation module for homogeneous nodes, and a training module under low resource conditions;

[0193] The general subgraph paradigm neighbor...

Embodiment 3

[0199] The present invention: discloses an electronic device. The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps in any one of the heterogeneous subgraph neural network-based similar information retrieval method in any one of the disclosed embodiments of the present invention are realized .

[0200] Figure 6 It is a structural diagram of an electronic device according to an embodiment of the present invention, such as Figure 6 As shown, the electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Wherein, the processor of the electronic device is used to provide calculation and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. ...

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 provides a similar information retrieval method and system based on an isoproton map neural network. The method comprises the following steps: firstly, carrying out graph structured data on a heterogeneous graph by using business scene data, namely constructing the heterogeneous graph; secondly, designing a sub-graph normal form, designing an isoproton graph neural network, modeling and learning neighborhood information of a center node according to the sub-graph normal form, and carrying out model training under the low-resource condition of not needing labels such as interactive records and the like, so as to obtain embedded representation of the node; and finally, designing a rapid similarity calculation module based on locality sensitive hashing to realize online service of a similar content retrieval function. According to the method and the device, the service requirement of similar information retrieval in a low-resource scene can be met.

Description

technical field [0001] The invention belongs to the field of similar information retrieval, and in particular relates to a similar information retrieval method and system based on a heterogeneous subgraph neural network. Background technique [0002] Similar content retrieval is a common and essential function in information retrieval systems, and it is extremely necessary in multiple business scenarios. For example, in e-commerce recommendation, it is necessary to search for products similar to the purchased products. Such products are considered to be in line with the user's historical purchase interests, thereby increasing the click-through rate and increasing transaction volume; Content, the most common method is to retrieve news similar to the news that the user has browsed. For example, a user has browsed the "Chinese Super League" news (soccer), and at this time needs to retrieve the "Champions League" related news (soccer); in web search, Often only limited content ...

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): G06F16/901G06F16/903G06F16/9535G06N3/04G06N3/08
CPCG06F16/9024G06F16/9535G06F16/90335G06N3/04G06N3/08
Inventor 陶建华槐泽鹏杨国花张大伟李冠君
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
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