Method for analyzing latent semantics of fusion probability of multi-modality data

A semantic analysis, multimodal technology, applied in the field of probabilistic latent semantic analysis, can solve the problems of data content expression and contribution difference, and achieve the effect of reducing workload and accurately analyzing results

Inactive Publication Date: 2011-12-21
BEIJING JIAOTONG UNIV
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Generally, probabilistic latent semantic analysis models are only suitable for single-modal data, and have certain limitations for more complex multi-modal data
Multimodal data contain multiple interrelated modalities that attempt to describe the same content but differ in their contribution to the expression of the data content

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
  • Method for analyzing latent semantics of fusion probability of multi-modality data
  • Method for analyzing latent semantics of fusion probability of multi-modality data
  • Method for analyzing latent semantics of fusion probability of multi-modality data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The preferred embodiments will be described in detail below in conjunction with the accompanying drawings. It should be emphasized that the following description is only exemplary and not intended to limit the scope of the invention and its application.

[0034] The technical problem to be solved by the present invention is to provide a probabilistic latent semantic analysis method for multi-modal data, which can simultaneously represent the difference in the contribution of each modality to the semantic space in the multi-modal data and the relationship between them Therefore, probabilistic latent semantic analysis can be reasonably extended from unimodal data to multimodal data, providing a more accurate description for it.

[0035] The present invention comprises the following steps:

[0036] Step 1: Establish a standard probabilistic latent semantic analysis model for each modality, and build a fusion model on this basis;

[0037] Step 2: Determine the workspace o...

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 method for analyzing latent semantics of the fusion probability of multi-modality data in the technical field of latent semantics analysis of probability. By starting from the essence of the multi-modality data and introducing a crossed structure, a standard probability latent semantic analysis model which is only applicable to single-modality is expanded to multi-modality, and difference between contributes of each modality to a latent semantic space and relevance of contents of each modality are modeled, so more accurate analysis and description are provided for the multi-modality data. The method has the advantages that: more accurate parameter estimation is realized by global parameter updating; reference is provided for selection of a proper subject number value range for each modality; and the working load of manual selection is reduced.

Description

technical field [0001] The invention belongs to the technical field of probabilistic latent semantic analysis, in particular to a fusion probabilistic latent semantic analysis method of multimodal data. Background technique [0002] Probabilistic Latent Semantic Analysis is a generation model of observations (word-document co-occurrence), which uses a mixture of multinomial distribution and conditional distribution to model the probability of co-occurrence through the introduction of latent semantic space. At present, it has been widely used in information retrieval, natural language processing, audio and video processing and other research fields. [0003] Usually, probabilistic latent semantic analysis models are only suitable for unimodal data, and have certain limitations for more complex multimodal data. Multimodal data contain multiple interrelated modalities that attempt to describe the same content but differ in their contribution to the expression of the data conte...

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): G06F17/27G06F17/30
Inventor 苗振江钟岑岑
Owner BEIJING JIAOTONG UNIV
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