Dimensionality Reduction with Sparse Locality for Principal Component Analysis
Various dimensionality reduction (DR) schemes have been developed for projecting high-dimensional data into low-dimensional representation. The existing schemes usually preserve either only the global structure or local structure of the original data, but not both. To resolve this issue, a scheme ca...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Hindawi Limited
2020-01-01
|
Series: | Mathematical Problems in Engineering |
Online Access: | http://dx.doi.org/10.1155/2020/9723279 |
id |
doaj-378d149c0bce422f80a92b9c68fc6fc3 |
---|---|
record_format |
Article |
spelling |
doaj-378d149c0bce422f80a92b9c68fc6fc32020-11-25T03:09:22ZengHindawi LimitedMathematical Problems in Engineering1024-123X1563-51472020-01-01202010.1155/2020/97232799723279Dimensionality Reduction with Sparse Locality for Principal Component AnalysisPei Heng Li0Taeho Lee1Hee Yong Youn2Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Republic of KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Republic of KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Republic of KoreaVarious dimensionality reduction (DR) schemes have been developed for projecting high-dimensional data into low-dimensional representation. The existing schemes usually preserve either only the global structure or local structure of the original data, but not both. To resolve this issue, a scheme called sparse locality for principal component analysis (SLPCA) is proposed. In order to effectively consider the trade-off between the complexity and efficiency, a robust L2,p-norm-based principal component analysis (R2P-PCA) is introduced for global DR, while sparse representation-based locality preserving projection (SR-LPP) is used for local DR. Sparse representation is also employed to construct the weighted matrix of the samples. Being parameter-free, this allows the construction of an intrinsic graph more robust against the noise. In addition, simultaneous learning of projection matrix and sparse similarity matrix is possible. Experimental results demonstrate that the proposed scheme consistently outperforms the existing schemes in terms of clustering accuracy and data reconstruction error.http://dx.doi.org/10.1155/2020/9723279 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Pei Heng Li Taeho Lee Hee Yong Youn |
spellingShingle |
Pei Heng Li Taeho Lee Hee Yong Youn Dimensionality Reduction with Sparse Locality for Principal Component Analysis Mathematical Problems in Engineering |
author_facet |
Pei Heng Li Taeho Lee Hee Yong Youn |
author_sort |
Pei Heng Li |
title |
Dimensionality Reduction with Sparse Locality for Principal Component Analysis |
title_short |
Dimensionality Reduction with Sparse Locality for Principal Component Analysis |
title_full |
Dimensionality Reduction with Sparse Locality for Principal Component Analysis |
title_fullStr |
Dimensionality Reduction with Sparse Locality for Principal Component Analysis |
title_full_unstemmed |
Dimensionality Reduction with Sparse Locality for Principal Component Analysis |
title_sort |
dimensionality reduction with sparse locality for principal component analysis |
publisher |
Hindawi Limited |
series |
Mathematical Problems in Engineering |
issn |
1024-123X 1563-5147 |
publishDate |
2020-01-01 |
description |
Various dimensionality reduction (DR) schemes have been developed for projecting high-dimensional data into low-dimensional representation. The existing schemes usually preserve either only the global structure or local structure of the original data, but not both. To resolve this issue, a scheme called sparse locality for principal component analysis (SLPCA) is proposed. In order to effectively consider the trade-off between the complexity and efficiency, a robust L2,p-norm-based principal component analysis (R2P-PCA) is introduced for global DR, while sparse representation-based locality preserving projection (SR-LPP) is used for local DR. Sparse representation is also employed to construct the weighted matrix of the samples. Being parameter-free, this allows the construction of an intrinsic graph more robust against the noise. In addition, simultaneous learning of projection matrix and sparse similarity matrix is possible. Experimental results demonstrate that the proposed scheme consistently outperforms the existing schemes in terms of clustering accuracy and data reconstruction error. |
url |
http://dx.doi.org/10.1155/2020/9723279 |
work_keys_str_mv |
AT peihengli dimensionalityreductionwithsparselocalityforprincipalcomponentanalysis AT taeholee dimensionalityreductionwithsparselocalityforprincipalcomponentanalysis AT heeyongyoun dimensionalityreductionwithsparselocalityforprincipalcomponentanalysis |
_version_ |
1715291974043435008 |