Clustering integration method based on weighted similarity measurement
A technology of weighted similarity and ensemble methods, applied in the field of cluster ensemble analysis
Inactive Publication Date: 2019-05-31
SHANXI UNIV
View PDF4 Cites 7 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The technical problem to be solved by the present invention is: to design a cluster integration method, carry out weighted similarity measurement according to the quality of cluster members, strengthen the positive influence of high-quality cluster members in the integration process, and suppress low-quality cluster members at the same time Unfavorable disturbances to obtain more accurate and robust clustering integration results
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
[0076] Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.
[0077] The cluster integration method based on the weighted similarity measure of the present invention is implemented by a computer program, figure 1 Shown is a computer-implemented system structure diagram. The technical scheme proposed by the present invention is used to process remote sensing image data below, remote sensing images are automatically classified, and the recognition of ground object targets is realized. The input data is an image data set composed of pixels, and the clustering integration method will have similar spectral characteristics. The pixel points are grouped into one category, identified as the same object, and finally the various objects identified are output. The specific implementation process is as follows: figure 2 shown. The spectral feature of each pixel in the remote sensing image is taken as a sample, an...
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 relates to the field of cluster integration analysis, in particular to a cluster integration method based on weighted similarity measurement. Weighted similarity measurement is carried out according to the quality of the clustering members, the positive influence of the high-quality clustering members is enhanced in the integration process, and meanwhile, the adverse interference ofthe low-quality clustering members is inhibited, so that a clustering integration result with higher accuracy and robustness is obtained. The method comprises the following steps: firstly, calculatingthe description consistency of any two samples in a data set on symbol space data in each clustering member; calculating the description consistency of each clustering member to the feature space data, and calculating the integration weight of each clustering member according to the characteristic space data description consistency; on the basis, calculating weighted similarity of any two samplesin the data set, then constructing a weighted similarity matrix of the data set so as to convert a clustering integration task into a graph minimum segmentation problem, obtaining a clustering integration result through solving by utilizing a spectral clustering method, and finally outputting the result.
Description
technical field [0001] The invention relates to the field of cluster integration analysis, in particular to a cluster integration method based on weighted similarity measure. Background technique [0002] Cluster analysis is an important and active research field in data mining. As an unsupervised learning method, clustering is essentially a density estimation problem. The data that needs to be clustered has not been labeled in advance and can be generated by a mixture model. Its main idea is to divide the data into several classes or clusters (groups), so that the similarity of data objects within a cluster is maximized and the similarity of data objects between clusters is minimized. In recent years, large-scale data sets have emerged frequently in various fields, which poses new challenges to cluster analysis research. In the face of large-scale data, traditional clustering analysis algorithms are no longer as "handy" as processing small and medium-scale data, but there...
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
Inventor 白亮杜航原
Owner SHANXI UNIV