Structure sparse multi-kernel learning-based multi-modal data feature screening and classification method

A data feature and multi-core learning technology, which is applied in character and pattern recognition, instruments, computer components, etc., to achieve the effect of simple and compact model and reduce model complexity

Active Publication Date: 2016-12-21
HUAQIAO UNIVERSITY
View PDF6 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The key technical problem to be solved by the present invention is: in the face of feature selection and classification of high-dimensional and multi-modal data, how to effectively reduce the data dimension so as to overcome the adverse effects of insufficient samples, and at the same time facilitate the establishment of a robust model that is easy to explain ; How to overcome the heterogeneous characteristics of different modal data, and how to use the different information of different modal data and

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
  • Structure sparse multi-kernel learning-based multi-modal data feature screening and classification method
  • Structure sparse multi-kernel learning-based multi-modal data feature screening and classification method
  • Structure sparse multi-kernel learning-based multi-modal data feature screening and classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041]The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

[0042] As a specific example, the data of senile dementia are adopted as follows. The data include nuclear magnetic resonance data (Nuclear Magnetic Resonance Imaging, MRI for short) and positron emission computed tomography data (Positron emission tomography, PET for short). The purpose is to use MRI Two types of medical imaging data, PET and PET, are used to diagnose whether an individual suffers from Alzheimer's disease, such as figure 1 What is shown describes the problems to be solved in this embodiment. The data contains image data of 90 individuals, and each individual data contains data of both MRI and PET modalities; in this data, there are 42 individual...

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 Structure sparse multi-kernel learning-based multi-modal data feature screening and classification method. The method comprises the following steps of: extracting data features and normalizing the features; constructing a kernel matrix for each data feature by utilizing a kernel function; grouping all the data features and the corresponding kernel functions according to the data features; training a Structure sparse multi-kernel classification model by utilizing training data with class labels, and optimizing parameters of the model; and classifying test data by using the trained Structure sparse multi-kernel classification model. According to the method, the data feature selection and data modal fusion are modeled in the uniform structure sparse multi-modal classification model, and the structure sparse feature selection and classifier learning expressed on the basis of an optimal kernel are carried out at the same time, so that a multi-modal data feature screening, fusion and classification method is provided.

Description

technical field [0001] The invention relates to a method for screening and classifying multimodal data features, in particular to a method for screening and classifying multimodal data features based on structural sparse multi-kernel learning, which belongs to the fields of machine learning and biomedicine, and is specifically applied to image and and / or in the field of disease understanding and diagnosis from genetic data. Background technique [0002] Many problems in our actual production and application can be attributed to classification problems. With the rapid increase of various data acquisition methods, we can often have data from multiple data sources (or called multi-modal). These different data describe different aspects of specific problems, but different modal data are heterogeneous, and how to effectively utilize multi-modal data to enhance classification becomes a major challenge. For example, old age is a neurodegenerative disease with insidious onset and ...

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): G06K9/62
CPCG06F18/2155G06F18/241
Inventor 彭佳林王烨王靖张洪博
Owner HUAQIAO UNIVERSITY
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