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

A streaming data processing method based on on-line multi-dimensional output of correlation

A technology of streaming data and processing methods, which is applied in the field of big data related analysis and machine learning, can solve problems such as consuming a lot of time and space resources, and achieve the effect of improving prediction performance and good use value

Inactive Publication Date: 2019-01-08
青岛海洋科学与技术试点国家实验室发展中心 +1
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in practical applications, if the data is too large to be processed at one time or the learned results have to be discarded only because of the arrival of new data, it will consume a lot of time and space resources to re-learn all the data.

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
  • A streaming data processing method based on on-line multi-dimensional output of correlation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will comprehensively use the basic methods and the latest research results of related disciplines such as big data correlation analysis, statistical learning theory, convex optimization theory, probability theory, machine learning and data mining, from the high coupling, mass and real-time nature of big data Based on the characteristics of the paper, and using the method of combining theoretical derivation and experimental analysis, an online multi-dimensional output support vector regression algorithm based on correlation is proposed. Predictive performance of multiple output regression.

[0036] The present invention is based on the online multidimensional output support vector regression method, including the following implementation process:

[0037] Define the correlation measure and correlation learning between the multidimensional weight vectors of the multidimensional regression estimation function;

[0038] Establish a multidimensional ou...

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

A streaming data processing method based on on-line multi-dimensional output of correlation is provided. Firstly, the correlation measure between the multidimensional output variables of the multidimensional regression estimation function is selected, streaming data is used to learn about relevance measures, the correlation between the multidimensional output variables of the multidimensional regression estimation function is obtained, then the regularization model of multi-dimensional output variable correlation is established, then the multidimensional output variable correlation learning system of multidimensional output variable is obtained, finally, based on the correlation learning system of multi-dimensional output variables, the correlation-based multi-output support vector regression model and its estimation function are obtained, and the multi-output support vector regression model is updated according to the new sample of the multi-dimensional output training data set.

Description

technical field [0001] The invention relates to the fields of big data correlation analysis and machine learning, in particular a correlation-based online multi-dimensional output streaming data processing method. Background technique [0002] At present, all walks of life have accumulated a large amount of data resources, how to effectively analyze various big data, and efficiently mine useful information hidden behind it from big data to predict the future development trend of things is the current information Research hotspots in the field of science. [0003] Regression estimation in multidimensional output learning, as an effective data mining and prediction technology, has been widely used in stock price prediction, emergency casualty prediction, fault diagnosis, channel estimation, air ticket price prediction, wind speed prediction, computer vision, etc. field. Multidimensional output regression aims to mine the functional dependencies between multidimensional input...

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
IPC IPC(8): G06F16/2458G06N20/10
CPCG06F2216/03
Inventor 王甫张玉强胡权威栾京东李金旭马琪杜鹏
Owner 青岛海洋科学与技术试点国家实验室发展中心
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