Unstructured view missing data classification method

An unstructured and missing data technology, applied in unstructured text data retrieval, text database clustering/classification, electronic digital data processing, etc., to achieve high efficiency, low complexity, and improve classification efficiency.

Inactive Publication Date: 2018-03-06
SOUTHWEST JIAOTONG UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0016] The purpose of the present invention is to provide a classification method for missing data in unstructured views, which can effectively solve the problem of extracting information from the original data, improving the accuracy of data analysis, reducing errors, and broadening the scope of practical applications

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
  • Unstructured view missing data classification method
  • Unstructured view missing data classification method
  • Unstructured view missing data classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The present invention will be described in further detail below in conjunction with the accompanying drawings.

[0055] A method for classifying missing data from unstructured views. A MMLE (Multi-task Multi-view Laplacian Eigenmaps) framework based on multi-task multi-view Laplacian Eigenmaps is proposed to enrich the information content of original data. The framework consists of two stages: Data processing in the absence of structured views, multi-task multi-view classification in the absence of unstructured views. It includes the following steps:

[0056] Step 1. Processing of unstructured missing data, such as figure 1 As shown, the processing steps are as follows:

[0057] (1) Construction dictionary B=[b 1 ,b 2 ,...,b i ,...,b p ]∈R m×p , there are three ways to construct dictionary B:

[0058] 1) Directly use the complete data set as the elements in the dictionary B;

[0059] 2) Select the K nearest neighbors of the unstructured missing data as the elem...

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 an unstructured view missing data classification method, and belongs to the technical field of data mining. The method comprises an unstructured view missing data processing stage and an unstructured view missing multi-task multi-view classification stage, wherein the unstructured view missing data processing stage comprises the following steps of: constructing a dictionary, solving a reconstruction error of complete data by utilizing local constraint sparse expression, solving a sample matrix and a dictionary after conversion, calculating distances between samples andthe dictionary, solving Lasso optimization and estimating a missing value; and the unstructured view missing multi-task multi-view classification stage comprises the following steps of: constructing asample graph, determining a weight, converting the samples to a feature mapping space, and classifying the samples by utilizing an MMLE framework. The method can be used for data missing in data transfer processes in practical scenes, extracting original data features to the greatest extent and keeping distributions and structures of data, thereby greatly enhancing the performance and applicationrange of multi-task multi-view classification.

Description

technical field [0001] The invention belongs to the technical field of data mining. Background technique [0002] Multi-task and multi-view learning has long been a hot research topic in the field of data mining, such as: multi-task and multi-view clustering problems, multi-task and multi-view classification integration problems, and so on. Then, most of the research topics based on multi-task and multi-view learning are carried out on complete data sets. The complete data set in the process of data processing and data analysis is often only the ideal state designed by researchers, and the complete data set in the actual scene Sets are often difficult to preserve. Therefore, research on missing data is imperative. [0003] However, missing data based on multiple views can be divided into two cases: missing structured views and missing unstructured views. Missing structured views means that all information under a view or a feature cannot be obtained; missing unstructured ...

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): G06F17/30
CPCG06F16/355G06F16/35
Inventor 杨燕张熠玲周威
Owner SOUTHWEST 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