Towards subspace clustering on dynamic data: An incremental version of PreDeCon

Hans Peter Kriegel*, Peer Kröger, Irene Ntoutsi, Arthur Zimek

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

Todays data are high dimensional and dynamic, thus clustering over such kind of data is rather complicated. To deal with the high dimensionality problem, the subspace clustering research area has lately emerged that aims at finding clusters in subspaces of the original feature space. So far, the subspace clustering methods are mainly static and thus, cannot address the dynamic nature of modern data. In this paper, we propose an incremental version of the density based projected clustering algorithm PreDeCon, called in-cPreDeCon. The proposed algorithm efficiently updates only those subspace clusters that might be affected due to the population update.

Original languageEnglish
Title of host publicationProceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Publication date25. Jul 2010
Pages31-38
ISBN (Print)978-1-4503-0226-5
DOIs
Publication statusPublished - 25. Jul 2010
Externally publishedYes
Event1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining - Washington, United States
Duration: 25. Jul 201025. Jul 2010

Conference

Conference1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
CountryUnited States
CityWashington
Period25/07/201025/07/2010
SponsorACM Spec. Interest Group Knowl. Discov. Data (SIGKDD), ACM Special Interest Group on Management of Data (SIGMOD)

Fingerprint

Clustering algorithms

Cite this

Kriegel, H. P., Kröger, P., Ntoutsi, I., & Zimek, A. (2010). Towards subspace clustering on dynamic data: An incremental version of PreDeCon. In Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 31-38). Association for Computing Machinery. https://doi.org/10.1145/1833280.1833285
Kriegel, Hans Peter ; Kröger, Peer ; Ntoutsi, Irene ; Zimek, Arthur. / Towards subspace clustering on dynamic data : An incremental version of PreDeCon. Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2010. pp. 31-38
@inproceedings{d9a48506f82b4962862d97ea06e00a1b,
title = "Towards subspace clustering on dynamic data: An incremental version of PreDeCon",
abstract = "Todays data are high dimensional and dynamic, thus clustering over such kind of data is rather complicated. To deal with the high dimensionality problem, the subspace clustering research area has lately emerged that aims at finding clusters in subspaces of the original feature space. So far, the subspace clustering methods are mainly static and thus, cannot address the dynamic nature of modern data. In this paper, we propose an incremental version of the density based projected clustering algorithm PreDeCon, called in-cPreDeCon. The proposed algorithm efficiently updates only those subspace clusters that might be affected due to the population update.",
author = "Kriegel, {Hans Peter} and Peer Kr{\"o}ger and Irene Ntoutsi and Arthur Zimek",
year = "2010",
month = "7",
day = "25",
doi = "10.1145/1833280.1833285",
language = "English",
isbn = "978-1-4503-0226-5",
pages = "31--38",
booktitle = "Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",
address = "United States",

}

Kriegel, HP, Kröger, P, Ntoutsi, I & Zimek, A 2010, Towards subspace clustering on dynamic data: An incremental version of PreDeCon. in Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, pp. 31-38, 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Washington, United States, 25/07/2010. https://doi.org/10.1145/1833280.1833285

Towards subspace clustering on dynamic data : An incremental version of PreDeCon. / Kriegel, Hans Peter; Kröger, Peer; Ntoutsi, Irene; Zimek, Arthur.

Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2010. p. 31-38.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - Towards subspace clustering on dynamic data

T2 - An incremental version of PreDeCon

AU - Kriegel, Hans Peter

AU - Kröger, Peer

AU - Ntoutsi, Irene

AU - Zimek, Arthur

PY - 2010/7/25

Y1 - 2010/7/25

N2 - Todays data are high dimensional and dynamic, thus clustering over such kind of data is rather complicated. To deal with the high dimensionality problem, the subspace clustering research area has lately emerged that aims at finding clusters in subspaces of the original feature space. So far, the subspace clustering methods are mainly static and thus, cannot address the dynamic nature of modern data. In this paper, we propose an incremental version of the density based projected clustering algorithm PreDeCon, called in-cPreDeCon. The proposed algorithm efficiently updates only those subspace clusters that might be affected due to the population update.

AB - Todays data are high dimensional and dynamic, thus clustering over such kind of data is rather complicated. To deal with the high dimensionality problem, the subspace clustering research area has lately emerged that aims at finding clusters in subspaces of the original feature space. So far, the subspace clustering methods are mainly static and thus, cannot address the dynamic nature of modern data. In this paper, we propose an incremental version of the density based projected clustering algorithm PreDeCon, called in-cPreDeCon. The proposed algorithm efficiently updates only those subspace clusters that might be affected due to the population update.

U2 - 10.1145/1833280.1833285

DO - 10.1145/1833280.1833285

M3 - Article in proceedings

AN - SCOPUS:77956233065

SN - 978-1-4503-0226-5

SP - 31

EP - 38

BT - Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -

Kriegel HP, Kröger P, Ntoutsi I, Zimek A. Towards subspace clustering on dynamic data: An incremental version of PreDeCon. In Proceedings of the 1st International Workshop on Novel Data Stream Pattern Mining Techniques, StreamKDD'10, Held in Conjunction with the 16th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2010. p. 31-38 https://doi.org/10.1145/1833280.1833285