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

Rating prediction method and system based on capsule network and interactive attention mechanism

An attention and capsule technology, applied in computer parts, instruments, text database query, etc., can solve problems such as information that cannot be further highlighted, and achieve the effect of improving interpretability and improving the accuracy of rating prediction.

Active Publication Date: 2022-07-05
QILU UNIV OF TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This leads to the weight of contextual features being gradually fixed during the learning process, which cannot further highlight aspect-related information

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
  • Rating prediction method and system based on capsule network and interactive attention mechanism
  • Rating prediction method and system based on capsule network and interactive attention mechanism
  • Rating prediction method and system based on capsule network and interactive attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] The purpose of this embodiment is an item recommendation method based on capsule network and interactive attention mechanism.

[0044] An item recommendation method based on capsule network and interactive attention mechanism, including:

[0045] Obtain user review data and item review data, and construct user documents and item documents respectively;

[0046] Inputting the user document and the item document into a pre-trained rating prediction model to obtain a user-item rating prediction result;

[0047] Wherein, the rating prediction model includes a content encoding unit, an interactive attention unit, a reverse dynamic routing unit and a prediction unit connected in sequence, and the content encoding unit extracts the contextual features of the user document and the item document respectively; through the interactive attention The force unit learns the fine-grained correlation between the contextual features of users and items; the reverse dynamic routing unit a...

Embodiment 2

[0113] The purpose of this embodiment is an item recommendation system based on capsule network and interactive attention mechanism.

[0114] An item recommendation system based on capsule network and interactive attention mechanism, including:

[0115] The data acquisition module is configured to acquire user comment data and item comment data, and construct user documents and item documents respectively;

[0116] a rating prediction module, configured to input the user document and the item document into a pre-trained rating prediction model to obtain a user-item rating prediction result;

[0117] Wherein, the rating prediction module includes:

[0118] a content encoding unit, configured to extract contextual features of the user document and the item document, respectively;

[0119] An interactive attention unit, configured to learn fine-grained correlations between contextual features of users and items;

[0120] The reverse dynamic routing unit is configured to aggreg...

Embodiment 3

[0123] The purpose of this embodiment is to provide an electronic device.

[0124] An electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the memory, the processor implements the described rating prediction method based on a capsule network and an interactive attention mechanism when the processor executes the program, including:

[0125] Obtain user review data and item review data, and construct user documents and item documents respectively;

[0126] Inputting the user document and the item document into a pre-trained rating prediction model to obtain a user-item rating prediction result;

[0127] Wherein, the rating prediction model includes a content encoding unit, an interactive attention unit, a reverse dynamic routing unit and a prediction unit connected in sequence, and the content encoding unit extracts the contextual features of the user document and the item document respectively; through the interactive a...

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 disclosure provides a rating prediction method and system based on a capsule network and an interactive attention mechanism. The solution integrates the contextual features of users and items and generates aspect features through a designed reverse dynamic routing with strong interpretability. At the same time, an interactive attention mechanism is proposed, through which the interaction between the user and item context features is constructed in multiple feature subspaces, so that the model can learn fine-grained interactive information, which effectively alleviates the problem of each Convergence among aspect features improves the accuracy of rating predictions.

Description

technical field [0001] The present disclosure belongs to the technical field of rating prediction, and in particular relates to a rating prediction method and system based on a capsule network and an interactive attention mechanism. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] As a valuable source of information for recommender systems, review texts usually contain rich semantics with user preferences and item attributes. In the prior art, the mainstream recommendation models use deep learning technology to model the review text. These methods use an association matrix to model the correlation between users and items, and then aggregate the contextual features of users and items to form a certain A potential representation of interpretability. The inventor found that the limitation of the existing methods is that the correlation matr...

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): G06F16/33G06F16/35G06F16/9535G06K9/62
CPCG06F16/3344G06F16/353G06F16/9535G06F18/214
Inventor 杨振宇刘国敬王皓
Owner QILU UNIV OF TECH
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