Instance Weighted Clustering: Local Outlier Factor and K-Means

Paul Moggridge; Na Helian; Yi Sun; Mariana Lilley; Vito Veneziano

doi:10.1007/978-3-030-48791-1_34

Instance Weighted Clustering: Local Outlier Factor and K-Means

Paul Moggridge, Na Helian, Yi Sun, Mariana Lilley, Vito Veneziano

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

54 Downloads (Pure)

Abstract

Clustering is an established unsupervised learning method. Substantial research has been carried out in the area of feature weighting, as well instance selection for clustering. Some work has paid attention to instance weighted clustering algorithms using various instance weighting metrics based on distance information, geometric information and entropy information. However, little research has made use of instance density information to weight instances. In this paper we use density to define instance weights. We propose two novel instance weighted clustering algorithms based on Local Outlier Factor and compare them against plain k-means and traditional instance selection.

Original language	English
Title of host publication	Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference
Subtitle of host publication	Proceedings of the EANN 2020
Editors	Lazaros Iliadis, Plamen Parvanov Angelov, Chrisina Jayne, Elias Pimenidis
Publisher	Springer Nature
Pages	435-446
Number of pages	12
ISBN (Electronic)	9783030487911
ISBN (Print)	9783030487904
DOIs	https://doi.org/10.1007/978-3-030-48791-1_34
Publication status	Published - 28 May 2020

Publication series

Name	Proceedings of the International Neural Networks Society
Publisher	Springer
ISSN (Print)	2661-8141
ISSN (Electronic)	2661-815X

Access to Document

10.1007/978-3-030-48791-1_34Licence: Unspecified

loiwreport2_final submission
© 2020 Springer-Verlag. The final publication is available at Springer via https://doi.org/10.1007/978-3-030-48791-1_34.
Accepted author manuscript, 460 KBLicence: Unspecified

Cite this

Moggridge, P., Helian, N., Sun, Y., Lilley, M., & Veneziano, V. (2020). Instance Weighted Clustering: Local Outlier Factor and K-Means. In L. Iliadis, P. P. Angelov, C. Jayne, & E. Pimenidis (Eds.), Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference: Proceedings of the EANN 2020 (pp. 435-446). (Proceedings of the International Neural Networks Society). Springer Nature. https://doi.org/10.1007/978-3-030-48791-1_34

Moggridge, Paul ; Helian, Na ; Sun, Yi et al. / Instance Weighted Clustering: Local Outlier Factor and K-Means. Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference: Proceedings of the EANN 2020. editor / Lazaros Iliadis ; Plamen Parvanov Angelov ; Chrisina Jayne ; Elias Pimenidis. Springer Nature, 2020. pp. 435-446 (Proceedings of the International Neural Networks Society).

@inproceedings{fd1a468b0a4f4c7696f9562fa5d6d601,

title = "Instance Weighted Clustering: Local Outlier Factor and K-Means",

abstract = "Clustering is an established unsupervised learning method. Substantial research has been carried out in the area of feature weighting, as well instance selection for clustering. Some work has paid attention to instance weighted clustering algorithms using various instance weighting metrics based on distance information, geometric information and entropy information. However, little research has made use of instance density information to weight instances. In this paper we use density to define instance weights. We propose two novel instance weighted clustering algorithms based on Local Outlier Factor and compare them against plain k-means and traditional instance selection.",

author = "Paul Moggridge and Na Helian and Yi Sun and Mariana Lilley and Vito Veneziano",

note = "{\textcopyright} 2020 Springer-Verlag. The final publication is available at Springer via https://doi.org/10.1007/978-3-030-48791-1_34. ",

year = "2020",

month = may,

day = "28",

doi = "10.1007/978-3-030-48791-1_34",

language = "English",

isbn = "9783030487904",

series = "Proceedings of the International Neural Networks Society",

publisher = "Springer Nature",

pages = "435--446",

editor = "Lazaros Iliadis and Angelov, {Plamen Parvanov} and Chrisina Jayne and Elias Pimenidis",

booktitle = "Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference",

address = "Netherlands",

}

Moggridge, P , Helian, N , Sun, Y, Lilley, M & Veneziano, V 2020, Instance Weighted Clustering: Local Outlier Factor and K-Means. in L Iliadis, PP Angelov, C Jayne & E Pimenidis (eds), Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference: Proceedings of the EANN 2020. Proceedings of the International Neural Networks Society, Springer Nature, pp. 435-446. https://doi.org/10.1007/978-3-030-48791-1_34

Instance Weighted Clustering: Local Outlier Factor and K-Means. / Moggridge, Paul ; Helian, Na ; Sun, Yi et al.
Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference: Proceedings of the EANN 2020. ed. / Lazaros Iliadis; Plamen Parvanov Angelov; Chrisina Jayne; Elias Pimenidis. Springer Nature, 2020. p. 435-446 (Proceedings of the International Neural Networks Society).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Instance Weighted Clustering: Local Outlier Factor and K-Means

AU - Moggridge, Paul

AU - Helian, Na

AU - Sun, Yi

AU - Lilley, Mariana

AU - Veneziano, Vito

PY - 2020/5/28

Y1 - 2020/5/28

N2 - Clustering is an established unsupervised learning method. Substantial research has been carried out in the area of feature weighting, as well instance selection for clustering. Some work has paid attention to instance weighted clustering algorithms using various instance weighting metrics based on distance information, geometric information and entropy information. However, little research has made use of instance density information to weight instances. In this paper we use density to define instance weights. We propose two novel instance weighted clustering algorithms based on Local Outlier Factor and compare them against plain k-means and traditional instance selection.

AB - Clustering is an established unsupervised learning method. Substantial research has been carried out in the area of feature weighting, as well instance selection for clustering. Some work has paid attention to instance weighted clustering algorithms using various instance weighting metrics based on distance information, geometric information and entropy information. However, little research has made use of instance density information to weight instances. In this paper we use density to define instance weights. We propose two novel instance weighted clustering algorithms based on Local Outlier Factor and compare them against plain k-means and traditional instance selection.

U2 - 10.1007/978-3-030-48791-1_34

DO - 10.1007/978-3-030-48791-1_34

M3 - Conference contribution

SN - 9783030487904

T3 - Proceedings of the International Neural Networks Society

SP - 435

EP - 446

BT - Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference

A2 - Iliadis, Lazaros

A2 - Angelov, Plamen Parvanov

A2 - Jayne, Chrisina

A2 - Pimenidis, Elias

PB - Springer Nature

ER -

Moggridge P , Helian N , Sun Y, Lilley M, Veneziano V. Instance Weighted Clustering: Local Outlier Factor and K-Means. In Iliadis L, Angelov PP, Jayne C, Pimenidis E, editors, Proceedings of the 21st EANN (Engineering Applications of Neural Networks) 2020 Conference: Proceedings of the EANN 2020. Springer Nature. 2020. p. 435-446. (Proceedings of the International Neural Networks Society). doi: 10.1007/978-3-030-48791-1_34

Instance Weighted Clustering: Local Outlier Factor and K-Means

Abstract

Publication series

Access to Document

Fingerprint

Cite this