Entropy Regularized Unsupervised Clustering Based on Maximum Correntropy Criterion and Adaptive Neighbors

Xinyu LI; Hui FAN; Jinglei LIU

doi:10.1587/transinf.2022EDL8054

Entropy Regularized Unsupervised Clustering Based on Maximum Correntropy Criterion and Adaptive Neighbors

Xinyu LI, Hui FAN, Jinglei LIU

Full Text Views

0

Cite this

Summary :

Constructing accurate similarity graph is an important process in graph-based clustering. However, traditional methods have three drawbacks, such as the inaccuracy of the similarity graph, the vulnerability to noise and outliers, and the need for additional discretization process. In order to eliminate these limitations, an entropy regularized unsupervised clustering based on maximum correntropy criterion and adaptive neighbors (ERMCC) is proposed. 1) Combining information entropy and adaptive neighbors to solve the trivial similarity distributions. And we introduce l₀-norm and spectral embedding to construct similarity graph with sparsity and strong segmentation ability. 2) Reducing the negative impact of non-Gaussian noise by reconstructing the error using correntropy. 3) The prediction label vector is directly obtained by calculating the sparse strongly connected components of the similarity graph Z, which avoids additional discretization process. Experiments are conducted on six typical datasets and the results showed the effectiveness of the method.

Publication: IEICE TRANSACTIONS on Information Vol.E106-D No.1 pp.82-85

Publication Date: 2023/01/01

Publicized: 2022/10/06

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2022EDL8054

Type of Manuscript: LETTER

Category: Artificial Intelligence, Data Mining

Authors

Xinyu LI
  Shandong Technology and Business University
Hui FAN
  Shandong Technology and Business University
Jinglei LIU
  Yantai University

Keyword

adaptive neighbors, entropy regularized, half-quadratic optimization, maximum correntropy criterion

Cite this

Copy

Xinyu LI, Hui FAN, Jinglei LIU, "Entropy Regularized Unsupervised Clustering Based on Maximum Correntropy Criterion and Adaptive Neighbors" in IEICE TRANSACTIONS on Information, vol. E106-D, no. 1, pp. 82-85, January 2023, doi: 10.1587/transinf.2022EDL8054.
Abstract: Constructing accurate similarity graph is an important process in graph-based clustering. However, traditional methods have three drawbacks, such as the inaccuracy of the similarity graph, the vulnerability to noise and outliers, and the need for additional discretization process. In order to eliminate these limitations, an entropy regularized unsupervised clustering based on maximum correntropy criterion and adaptive neighbors (ERMCC) is proposed. 1) Combining information entropy and adaptive neighbors to solve the trivial similarity distributions. And we introduce l₀-norm and spectral embedding to construct similarity graph with sparsity and strong segmentation ability. 2) Reducing the negative impact of non-Gaussian noise by reconstructing the error using correntropy. 3) The prediction label vector is directly obtained by calculating the sparse strongly connected components of the similarity graph Z, which avoids additional discretization process. Experiments are conducted on six typical datasets and the results showed the effectiveness of the method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022EDL8054/_p

Copy

@ARTICLE{e106-d_1_82,
author={Xinyu LI, Hui FAN, Jinglei LIU, },
journal={IEICE TRANSACTIONS on Information},
title={Entropy Regularized Unsupervised Clustering Based on Maximum Correntropy Criterion and Adaptive Neighbors},
year={2023},
volume={E106-D},
number={1},
pages={82-85},
abstract={Constructing accurate similarity graph is an important process in graph-based clustering. However, traditional methods have three drawbacks, such as the inaccuracy of the similarity graph, the vulnerability to noise and outliers, and the need for additional discretization process. In order to eliminate these limitations, an entropy regularized unsupervised clustering based on maximum correntropy criterion and adaptive neighbors (ERMCC) is proposed. 1) Combining information entropy and adaptive neighbors to solve the trivial similarity distributions. And we introduce l₀-norm and spectral embedding to construct similarity graph with sparsity and strong segmentation ability. 2) Reducing the negative impact of non-Gaussian noise by reconstructing the error using correntropy. 3) The prediction label vector is directly obtained by calculating the sparse strongly connected components of the similarity graph Z, which avoids additional discretization process. Experiments are conducted on six typical datasets and the results showed the effectiveness of the method.},
keywords={},
doi={10.1587/transinf.2022EDL8054},
ISSN={1745-1361},
month={January},}

Copy

TY - JOUR
TI - Entropy Regularized Unsupervised Clustering Based on Maximum Correntropy Criterion and Adaptive Neighbors
T2 - IEICE TRANSACTIONS on Information
SP - 82
EP - 85
AU - Xinyu LI
AU - Hui FAN
AU - Jinglei LIU
PY - 2023
DO - 10.1587/transinf.2022EDL8054
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2023
AB - Constructing accurate similarity graph is an important process in graph-based clustering. However, traditional methods have three drawbacks, such as the inaccuracy of the similarity graph, the vulnerability to noise and outliers, and the need for additional discretization process. In order to eliminate these limitations, an entropy regularized unsupervised clustering based on maximum correntropy criterion and adaptive neighbors (ERMCC) is proposed. 1) Combining information entropy and adaptive neighbors to solve the trivial similarity distributions. And we introduce l₀-norm and spectral embedding to construct similarity graph with sparsity and strong segmentation ability. 2) Reducing the negative impact of non-Gaussian noise by reconstructing the error using correntropy. 3) The prediction label vector is directly obtained by calculating the sparse strongly connected components of the similarity graph Z, which avoids additional discretization process. Experiments are conducted on six typical datasets and the results showed the effectiveness of the method.
ER -