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

Social network cross-media search method based on adversarial learning and semantic similarity

A technology of semantic similarity and social network, applied in the fields of image processing and text processing, it can solve the problems of different distribution of data features, inability to establish association relationships, and inability to communicate in semantic space, so as to maintain appropriate semantics, facilitate subsequent training, and speed up features. The effect of extraction speed and accuracy

Inactive Publication Date: 2019-11-26
BEIJING UNIV OF POSTS & TELECOMM
View PDF3 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the distribution of data features between different modalities is different, and the semantic space cannot communicate with each other, that is, text and images containing the same semantic content cannot directly establish an association relationship through their semantic space.

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
  • Social network cross-media search method based on adversarial learning and semantic similarity
  • Social network cross-media search method based on adversarial learning and semantic similarity
  • Social network cross-media search method based on adversarial learning and semantic similarity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The present invention will be described in further detail below with reference to the accompanying drawings. The concrete realization of the algorithm of the present invention is divided into the following steps:

[0020] 1. Feature Mapping Network

[0021] The feature mapping network is divided into two parts, the image feature mapping network and the text feature mapping network, which are responsible for mapping the original data features into the same semantic space. In order to ensure that the mapped data maintains the semantic features of the original modality, a semantic prediction network is added behind the feature mapping network, and the output of the classifier softmax is used as a result to predict the semantic distribution of data mapped to the same semantic space. Let the parameters of the network be θ imd , the c-th dimension value of the i-th data semantic distribution in the two modalities of image and text is p ic (v i ) and p ic (t i ), use the...

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 social network cross-media search method based on adversarial learning and semantic similarity. A text and image feature extraction network, a public semantic space mapping network, a semantic similarity network and a modal discrimination network are included. The method has outstanding innovativeness, and is mainly used in social network cross-media search. The method isapplied to the field of image and text processing, cross-media data in different modes can be processed, and retrieval between the cross-media data is efficient and accurate.

Description

technical field [0001] The invention belongs to the technical field of image processing and text processing, and specifically relates to mutual retrieval between cross-media data, integrating various technologies, such as confrontational learning, deep neural network, semantic fusion, sorting search algorithm, etc., and finally realizing cross-media data Semantic association and search. Background technique [0002] With the rapid development of the mobile Internet, social network data information has shown explosive growth, and more and more users have released a large amount of real-time information on various social media, among which information related to national security is particularly important. When an accident or disaster occurs, texts and images related to the disaster can be searched in time, which can reduce the losses caused by the disaster to a certain extent. Microblog is an important part of social network, with the characteristics of short content, fast d...

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 Applications(China)
IPC IPC(8): G06F17/27G06N3/04G06Q50/00
CPCG06Q50/01G06N3/045
Inventor 杜军平薛哲刘翀周南
Owner BEIJING UNIV OF POSTS & TELECOMM
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