Method for solving social recommendation problem by low-rank semi-definite programming

A semi-definite programming and problem technology, applied in the field of using semi-definite programming to solve social recommendation problems, can solve problems such as complex algorithms and low accuracy, and achieve the effect of improving accuracy and increasing application value.

Active Publication Date: 2013-04-17
ZHEJIANG UNIV
View PDF3 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention overcomes the disadvantages of low precision and complex algorithms in existing social recommendation methods, and provides a method for solving social recommendation problems using semidefinite programming that guarantees recommendation accuracy and low computational complexity

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 solving social recommendation problem by low-rank semi-definite programming
  • Method for solving social recommendation problem by low-rank semi-definite programming
  • Method for solving social recommendation problem by low-rank semi-definite programming

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] figure 1 It is a flow chart of a preferred embodiment of a method for solving a social recommendation problem using a low-rank semidefinite programming of the present invention. The detailed implementation steps are as follows:

[0023] 1. In a recommender system with m users and n items, we extract the user's rating matrix M ∈ R m*n , then use Normalize the scores in M ​​to (0-1) to obtain a new user scoring matrix M′, where, are all rating values ​​[1,2…,r max ] mean;

[0024] 2. First, use the user scoring matrix M' to obtain the objective function of minimizing the difference

[0025] min U , V = | | I ⊗ ( M ′ - ρ ( U ...

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 solving a social recommendation problem by low-rank semi-definite programming. In the method, potential social relationships among users are investigated by graph-based laplacian regularization, and the problem is transformed into a low-rank semi-definite programming problem, so that the social recommendation problem can be efficiently handled by a quasi-Newton algorithm. In other words, when the graph-based laplacian regularization problem is handled, without a traditional gradient descent method, the problem is transformed into the low-rank semi-definite programming problem capable of being efficiently solved by the quasi-Newton algorithm. The method has the advantages that firstly, potential subspaces related to the users can be effectively handled by graph-based laplacian regularization to catch the potential relationships among the users, and secondly, the graph-based laplacian regularization problem can be directly transformed into the low-rank semi-definite programming problem which is easier to solve.

Description

technical field [0001] The invention relates to the fields of information retrieval, data mining, user modeling and the like, in particular to a method for solving social recommendation problems by using semi-definite programming. Background technique [0002] With the rapid development of the Internet and the maturation of Web 2.0 technology, recommendation systems play an increasingly important role in filtering a large amount of useless information for end users. At present, recommendation technology is not only a hot research topic, but also has huge potential commercial value in real life. In the past decade, recommendation techniques have been extensively studied in the fields of information retrieval and data mining. Most of the recommendation systems in the past are based on collaborative filtering technology, that is, predict the preferences of the current user through the information of other users. But in general, the items rated by users only account for a smal...

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
Inventor 朱建科卜佳俊陈纯王峰伟
Owner ZHEJIANG 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