An incomplete multi-view data clustering method and electronic equipment

A multi-view and complete technology, applied in the field of data analysis, can solve the problems of assigning weights to the importance of different views, the inability to accurately cluster multi-view data, and the lack of simultaneous utilization of global structural information and local structural information to achieve reliable clustering Results, the effect of improving clustering performance
CN113705603APending Publication Date: 2021-11-26BEIJING UNIV OF POSTS & TELECOMM +1
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN ยท China
Current Assignee / Owner
BEIJING UNIV OF POSTS & TELECOMM
Publication Date
2021-11-26

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides an incomplete multi-view data clustering method and electronic equipment, and the method comprises the steps of complementing missing multi-view features of incomplete multi-view data through a multi-view auto-encoder, so as to obtain complete multi-view data and unified feature representation thereof; learning a local structure of the complete multi-view data through a single-layer neural network model, and extracting local structure information of the complete multi-view data by using a graph convolutional network to obtain node feature representation of each view of the complete multi-view data; and based on the unified feature representation and the node feature representation, performing clustering through a preset clustering algorithm to obtain a clustering result of the complete multi-view data. According to the technical scheme, after missing features of incomplete multi-view data are complemented, feature representation of the multi-view data is enhanced by combining the global structure and the local structure of the multi-view data, and then a more accurate clustering result of the multi-view data is obtained.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present disclosure relates to the technical field of data analysis, and in particular to a clustering method and electronic equipment for incomplete multi-view data. Background technique

[0002] The existing clustering methods for incomplete multi-view data generally use deep multi-view autoencoders to learn a unified data representation for data from multiple views, and establish a set of multi-view autoencoders for the features of each view. Including the encoder part and the decoder part. For incomplete multi-view data, a weighted fusion method is used to fuse the outputs of each view encoder and represent them in a unified manner. At the same time, graph embedding constraints are added to the unified representation learning process so that the learned representation can retain local structural information. In addition, a clustering loss function is added after the unified representation layer to cluster multi-view data.

[0003] The traditio...

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