Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Matrix decomposition parallelization method based on graph calculation model

A matrix decomposition and graph calculation technology, applied in the field of recommendation systems, can solve problems such as suspended animation, huge messages, and crashes

Inactive Publication Date: 2016-09-28
ZHEJIANG SCI-TECH UNIV
View PDF3 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

That is, for a vertex with a large number of neighbors, the messages it needs to process are very large, and in this mode, they cannot be processed concurrently
Therefore, for a natural graph that conforms to a power law distribution, it is easy to freeze or crash under this calculation model

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
  • Matrix decomposition parallelization method based on graph calculation model
  • Matrix decomposition parallelization method based on graph calculation model
  • Matrix decomposition parallelization method based on graph calculation model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0080] Example: such as figure 1 Shown is a rating file for a user team product, where the UID column represents the user, and the IID column represents the product. Score represents the rating of the product by the corresponding user. A 4*3 scoring matrix can be constructed from the scoring file, such as figure 2 shown. The resulting scoring matrix can be generated in the form of a bipartite graph, such as image 3 shown. image 3 A bipartite graph is generated according to the scoring matrix. Each vertex in the bipartite graph represents a user or a product, and the user's rating of the product is the attribute of the edge between the user vertex and the product vertex. Therefore, two learning optimization algorithms based on coordinated filtering and matrix decomposition can be realized in the form of bipartite graphs.

[0081] Such as Figure 4 As shown, the solid line represents the actual interaction, and the dotted line represents the sending of the message. It c...

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 matrix decomposition parallelization method based on a graph calculation model. Matrix decomposition can be flexibly brought into more user information. The matrix decomposition deduces the hidden semantic vectors of a user and an article according to the score of the article by the user, and then, recommendation is carried out according to the hidden semantic vectors of the user and the article. However, in a practical application scene, the implementation of a matrix decomposition recommendation algorithm needs to consume a great quantity of time, and traditional commercial requirements can not be met. A distributed calculation platform can be used for carrying out parallelization on the matrix decomposition recommendation algorithm to effectively solve the problem, and meanwhile, a multiple-iteration calculation problem is in the presence in the implementation of the matrix decomposition recommendation algorithm. The invention puts forward the Spark-based GraphX graph calculation frame to realize matrix decomposition parallelization. Compared with a traditional MapReduce calculation graph model, the graph calculation frame has the obvious advantages on the aspect of the solving of multiple-iteration problems and execution efficiency.

Description

technical field [0001] The invention relates to the technical field of recommendation systems, in particular to a matrix decomposition parallelization method based on a graph computing model. Background technique [0002] In recent years, with the rapid development and popularization of computer and information technology, the scale of industrial application systems has expanded rapidly, and the data generated by industrial applications has grown explosively. Industries with a scale of hundreds of terabytes or even tens to hundreds of petabytes have far exceeded the processing capabilities of existing traditional computer technology and information systems. Therefore, it has become a reality to seek effective big data processing technologies, methods and means urgent needs of the world. The current total data volume of Baidu has exceeded 1000PB, and the webpage data that needs to be processed every day reaches 10PB~100PB; the cumulative transaction data volume of Taobao is ...

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): G06F17/16
CPCG06F17/16
Inventor 张娜戴世超包晓安熊子健
Owner ZHEJIANG SCI-TECH 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
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