Home > Research > Publications & Outputs > Weighted Sparse Simplex Representation

Research

Associated organisational units

Electronic data

  • Peng_Pavlidis2022

    Final published version, 730 KB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

Keywords

View graph of relations

Weighted Sparse Simplex Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Weighted Sparse Simplex Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning. / Peng, Hankui; Pavlidis, Nicos.
In: Data Mining and Knowledge Discovery, Vol. 36, No. 3, 31.05.2022, p. 958–986.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Peng, H & Pavlidis, N 2022, 'Weighted Sparse Simplex Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning', Data Mining and Knowledge Discovery, vol. 36, no. 3, pp. 958–986. https://doi.org/10.1007/s10618-022-00820-9

APA

Peng, H., & Pavlidis, N. (2022). Weighted Sparse Simplex Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning. Data Mining and Knowledge Discovery, 36(3), 958–986. https://doi.org/10.1007/s10618-022-00820-9

Vancouver

Peng H, Pavlidis N. Weighted Sparse Simplex Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning. Data Mining and Knowledge Discovery. 2022 May 31;36(3):958–986. Epub 2022 Feb 11. doi: 10.1007/s10618-022-00820-9

Author

Peng, Hankui ; Pavlidis, Nicos. / Weighted Sparse Simplex Representation : A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning. In: Data Mining and Knowledge Discovery. 2022 ; Vol. 36, No. 3. pp. 958–986.

Bibtex

@article{32085f7ab75745bfb4374f31b8f482ef,
title = "Weighted Sparse Simplex Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning",
abstract = "Spectral-based subspace clustering methods have proved successful in many challenging applications such as gene sequencing, image recognition, and motion segmentation. In this work, we first propose a novel spectral-based subspace clustering algorithm that seeks to represent each point as a sparse convex combination of a few nearby points. We then extend the algorithm to a constrained clustering and active learning framework. Our motivation for developing such a framework stems from the fact that typically either a small amount of labelled data are available in advance; or it is possible to label some points at a cost. The latter scenario is typically encountered in the process of validating a cluster assignment. Extensive experiments on simulated and real datasets show that the proposed approach is effective and competitive with state-of-the-art methods.",
keywords = "Subspace clustering, Constrained clustering, Active learning",
author = "Hankui Peng and Nicos Pavlidis",
year = "2022",
month = may,
day = "31",
doi = "10.1007/s10618-022-00820-9",
language = "English",
volume = "36",
pages = "958–986",
journal = "Data Mining and Knowledge Discovery",
issn = "1384-5810",
publisher = "Springer New York LLC",
number = "3",

}

RIS

TY - JOUR

T1 - Weighted Sparse Simplex Representation

T2 - A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning

AU - Peng, Hankui

AU - Pavlidis, Nicos

PY - 2022/5/31

Y1 - 2022/5/31

N2 - Spectral-based subspace clustering methods have proved successful in many challenging applications such as gene sequencing, image recognition, and motion segmentation. In this work, we first propose a novel spectral-based subspace clustering algorithm that seeks to represent each point as a sparse convex combination of a few nearby points. We then extend the algorithm to a constrained clustering and active learning framework. Our motivation for developing such a framework stems from the fact that typically either a small amount of labelled data are available in advance; or it is possible to label some points at a cost. The latter scenario is typically encountered in the process of validating a cluster assignment. Extensive experiments on simulated and real datasets show that the proposed approach is effective and competitive with state-of-the-art methods.

AB - Spectral-based subspace clustering methods have proved successful in many challenging applications such as gene sequencing, image recognition, and motion segmentation. In this work, we first propose a novel spectral-based subspace clustering algorithm that seeks to represent each point as a sparse convex combination of a few nearby points. We then extend the algorithm to a constrained clustering and active learning framework. Our motivation for developing such a framework stems from the fact that typically either a small amount of labelled data are available in advance; or it is possible to label some points at a cost. The latter scenario is typically encountered in the process of validating a cluster assignment. Extensive experiments on simulated and real datasets show that the proposed approach is effective and competitive with state-of-the-art methods.

KW - Subspace clustering

KW - Constrained clustering

KW - Active learning

U2 - 10.1007/s10618-022-00820-9

DO - 10.1007/s10618-022-00820-9

M3 - Journal article

VL - 36

SP - 958

EP - 986

JO - Data Mining and Knowledge Discovery

JF - Data Mining and Knowledge Discovery

SN - 1384-5810

IS - 3

ER -