Clustering method for dual regularization non-negative matrix factorization based on EMD measurement
A technique of non-negative matrix decomposition and clustering method, applied in the clustering field of dual regularization non-negative matrix decomposition, can solve problems such as linking together, unable to measure sample distance well, and achieve the effect of improving performance
Pending Publication Date: 2020-10-09
JIANGSU UNIV OF TECH
View PDF1 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
On the other hand, the traditional L2 distance cannot measure the distance between samples well due to the possible correlation between features
While some work has been proposed to address these issues, very little has been done linking them together
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0021] In order to deepen the understanding of the present invention, the present invention will be further described below in conjunction with the examples, which are only used to explain the present invention, and do not constitute a limitation to the protection scope of the present invention.
[0022] Such as figure 1 As shown, the present embodiment provides a clustering method based on the dual regularization non-negative matrix factorization of the EMD metric, the method comprising the following steps:
[0023] Step 1: Obtain the sample data to be clustered;
[0024] Step 2: Construct the adjacency matrix of the data manifold graph and the adjacency matrix of the feature manifold graph for the samples to be clustered;
[0025] Step 3: Obtain the objective function of the dual regularization non-negative matrix factorization based on the EMD metric through the regularization term of the data manifold graph and the regularization term of the feature manifold graph;
[00...
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
Login to View More Abstract
The invention provides a clustering method for dual regularization non-negative matrix factorization based on EMD measurement. The method comprises the following steps: 1, obtaining sample data to beclustered; 2, constructing an adjacency matrix of a data manifold graph and an adjacency matrix of a feature manifold graph of a sample to be clustered; 3, obtaining an objective function of dual regularization non-negative matrix factorization based on EMD measurement through a data manifold graph regularization item and a feature manifold graph regularization item; 4, setting the number of iterations by using an iterative weighting method according to the target function, and iteratively updating a coefficient matrix and a basis matrix in the NMF; and 5, clustering the iteratively updated data samples by adopting a k-means clustering algorithm. The performance of the NMF is improved by utilizing the information amount of the geometric structure, and the distance between samples is bettermeasured by adopting an EMD measurement mode.
Description
technical field [0001] The invention relates to the technical field of data processing, in particular to a clustering method of dual regularization non-negative matrix decomposition based on EMD measure. Background technique [0002] In recent years, high-dimensional data has appeared in many fields, and its dimensionality reduction operation has attracted people's attention. Non-negative matrix factorization (NMF), as a commonly used dimensionality reduction method, aims to learn local-based feature representations, and has been widely used in various applied research. Clustering is a basic topic in machine learning and data mining, the purpose is to divide a set of data into several groups according to the similarity of data points. Nonnegative matrix factorization (NMF) has received considerable attention due to its potential part-based in the human brain due to its psychological and physiological interpretation of naturally occurring data. Although NMF has good practic...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62
CPCG06F18/23213
Inventor 舒振球张云猛翁宗慧叶飞跃
Owner JIANGSU UNIV OF TECH