Dimensionality reduction via preserving local information

Wang, Shangguang; Ding, Chuntao; Hsu, Ching‐Hsien; Yang, Fangchun

doi:10.1016/j.future.2018.01.016

Cited by 12 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To demonstrate the effectiveness and feasibility of SBDP, it is chosen to compare with dimensionality reduction methods, such as supervised orthogonal local Fisher discriminant analysis (SOLFDA) [31], local similarity preserving discriminant (LSPD) [32], maximum nonparametric margin projection (MNMP) [33], MFA, UDP, LDP and ODP. Figure 6 shows the training results of different methods.…”

Section: Feature Dimensionality Reduction and Fault Identificationmentioning

confidence: 99%

Dimensionality reduction method based on similarity balance discriminant projection for bearing fault diagnosis

Shi¹,

Ding²,

Shi³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

For data-driven intelligent diagnosis, comprehensive mining of the information in the data is a key issue and a great challenge. The knowledge of how to discover potentially useful information in the data is particularly critical to raise the accuracy of fault pattern recognition. The structural and discriminative information in the data present a potential direction to solve the issue and overcome the challenge. In this study, a Similarity Balance Discriminant Projection (SBDP) algorithm is proposed, and by incorporating the optimized Support Vector Machine (SVM), and the develop a fault diagnosis model for rolling bearing fault diagnosis. To address the defects of unbalanced adjacency relations and non-strictly monotonic between-class weight functions, SBDP reconstructs new within-class adjacency graphs and between-class adjacency graphs by fusing structural information and category information, thus effectively preserving the intrinsic manifold structure information of high-dimensional feature data. The effectiveness and suitability of the proposed model is demonstrated by two kinds of bearing diagnosis applications. The application results indicate that SBDP is better able to extract features representing the intrinsic information of faults, the optimized SVM is effective in identifying the fault types with high accuracy.

show abstract

Section: Feature Dimensionality Reduction and Fault Identificationmentioning

confidence: 99%

Dimensionality reduction method based on similarity balance discriminant projection for bearing fault diagnosis

Shi¹,

Ding²,

Shi³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…The seventh paper "Dimensionality Reduction via Preserving Local Information" by Wang et al [9] addressed the problem of dimensionality reduction. For this purpose, they presented an effective and efficient metric function, named local similarity preserving (LSP), which can preserve the similarity information of each sample to its homogeneous and heterogeneous neighbors.…”

Section: Content Of This Special Issuementioning

confidence: 99%

Towards smarter cities: Learning from Internet of Multimedia Things-generated big data

Bellavista

Ota

et al. 2020

Future Generation Computer Systems

View full text Add to dashboard Cite

“…us, Ding et al constructed double adjacency graphs to link their homogeneous and heterogeneous neighbors and introduced a more effective version of DNE termed DAG-DNE [22]. Inspired by DAG-DNE, some discriminant analysis-based methods have been proposed over the past few years [23][24][25][26][27][28][29][30][31][32][33].…”

Section: Introductionmentioning

confidence: 99%

Large Margin Graph Embedding-Based Discriminant Dimensionality Reduction

Tian

Feng

2021

Scientific Programming

View full text Add to dashboard Cite

Discriminant graph embedding-based dimensionality reduction methods have attracted more and more attention over the past few decades. These methods construct an intrinsic graph and penalty graph to preserve the intrinsic geometry structures of intraclass samples and separate the interclass samples. However, the marginal samples cannot be accurately characterized only by penalty graphs since they treat every sample equally. In practice, these marginal samples often influence the classification performance, which needs to be specially tackled. In this study, the near neighbors’ hypothesis margin of marginal samples has been further maximized to separate the interclass samples and improve the discriminant ability by integrating intrinsic graph and penalty graph. A novel discriminant dimensionality reduction named LMGE-DDR has been proposed. Several experiments on public datasets have been conducted to verify the effectiveness of the proposed LMGE-DDR such as ORL, Yale, UMIST, FERET, CMIU-PIE09, and AR. LMGE-DDR performs better than other compared methods, and the corresponding standard deviation of LMGE-DDR is smaller than others. This demonstrates that the evaluation method verifies the effectiveness of the introduced method.

show abstract

Dimensionality reduction via preserving local information

Cited by 12 publications

References 19 publications

Dimensionality reduction method based on similarity balance discriminant projection for bearing fault diagnosis

Dimensionality reduction method based on similarity balance discriminant projection for bearing fault diagnosis

Towards smarter cities: Learning from Internet of Multimedia Things-generated big data

Large Margin Graph Embedding-Based Discriminant Dimensionality Reduction

Contact Info

Product

Resources

About