COSMIC: Conceptually specified multi-instance clusters

Hans Peter Kriegel*, Alexey Pryakhin, Matthias Schubert, Arthur Zimek

*Corresponding author for this work

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

Abstract

Recently, more and more applications represent data objects as sets of feature vectors or multi-instance objects. In this paper, we propose COSMIC, a method for deriving concept lattices from multi-instance data based on hierarchical density-based clustering. The found concepts correspond to groups or clusters of multi-instance objects having similar instances in common. We demonstrate that COSMIC outperforms compared methods with respect to efficiency and cluster quality and is capable to extract interesting patterns in multi-instance data sets.

Original languageEnglish
Title of host publicationProceedings - Sixth International Conference on Data Mining, ICDM 2006
PublisherIEEE
Publication dateDec 2006
Pages917-921
ISBN (Print)978-0-7695-2701-7
DOIs
Publication statusPublished - Dec 2006
Externally publishedYes
Event6th International Conference on Data Mining, ICDM 2006 - Hong Kong, China
Duration: 18. Dec 200622. Dec 2006

Conference

Conference6th International Conference on Data Mining, ICDM 2006
CountryChina
CityHong Kong
Period18/12/200622/12/2006

Cite this

Kriegel, H. P., Pryakhin, A., Schubert, M., & Zimek, A. (2006). COSMIC: Conceptually specified multi-instance clusters. In Proceedings - Sixth International Conference on Data Mining, ICDM 2006 (pp. 917-921). IEEE. https://doi.org/10.1109/ICDM.2006.46
Kriegel, Hans Peter ; Pryakhin, Alexey ; Schubert, Matthias ; Zimek, Arthur. / COSMIC : Conceptually specified multi-instance clusters. Proceedings - Sixth International Conference on Data Mining, ICDM 2006. IEEE, 2006. pp. 917-921
@inproceedings{e02d511a0fc14a09813bfb91201ab7fc,
title = "COSMIC: Conceptually specified multi-instance clusters",
abstract = "Recently, more and more applications represent data objects as sets of feature vectors or multi-instance objects. In this paper, we propose COSMIC, a method for deriving concept lattices from multi-instance data based on hierarchical density-based clustering. The found concepts correspond to groups or clusters of multi-instance objects having similar instances in common. We demonstrate that COSMIC outperforms compared methods with respect to efficiency and cluster quality and is capable to extract interesting patterns in multi-instance data sets.",
author = "Kriegel, {Hans Peter} and Alexey Pryakhin and Matthias Schubert and Arthur Zimek",
year = "2006",
month = "12",
doi = "10.1109/ICDM.2006.46",
language = "English",
isbn = "978-0-7695-2701-7",
pages = "917--921",
booktitle = "Proceedings - Sixth International Conference on Data Mining, ICDM 2006",
publisher = "IEEE",
address = "United States",

}

Kriegel, HP, Pryakhin, A, Schubert, M & Zimek, A 2006, COSMIC: Conceptually specified multi-instance clusters. in Proceedings - Sixth International Conference on Data Mining, ICDM 2006. IEEE, pp. 917-921, 6th International Conference on Data Mining, ICDM 2006, Hong Kong, China, 18/12/2006. https://doi.org/10.1109/ICDM.2006.46

COSMIC : Conceptually specified multi-instance clusters. / Kriegel, Hans Peter; Pryakhin, Alexey; Schubert, Matthias; Zimek, Arthur.

Proceedings - Sixth International Conference on Data Mining, ICDM 2006. IEEE, 2006. p. 917-921.

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

TY - GEN

T1 - COSMIC

T2 - Conceptually specified multi-instance clusters

AU - Kriegel, Hans Peter

AU - Pryakhin, Alexey

AU - Schubert, Matthias

AU - Zimek, Arthur

PY - 2006/12

Y1 - 2006/12

N2 - Recently, more and more applications represent data objects as sets of feature vectors or multi-instance objects. In this paper, we propose COSMIC, a method for deriving concept lattices from multi-instance data based on hierarchical density-based clustering. The found concepts correspond to groups or clusters of multi-instance objects having similar instances in common. We demonstrate that COSMIC outperforms compared methods with respect to efficiency and cluster quality and is capable to extract interesting patterns in multi-instance data sets.

AB - Recently, more and more applications represent data objects as sets of feature vectors or multi-instance objects. In this paper, we propose COSMIC, a method for deriving concept lattices from multi-instance data based on hierarchical density-based clustering. The found concepts correspond to groups or clusters of multi-instance objects having similar instances in common. We demonstrate that COSMIC outperforms compared methods with respect to efficiency and cluster quality and is capable to extract interesting patterns in multi-instance data sets.

U2 - 10.1109/ICDM.2006.46

DO - 10.1109/ICDM.2006.46

M3 - Article in proceedings

AN - SCOPUS:62649092399

SN - 978-0-7695-2701-7

SP - 917

EP - 921

BT - Proceedings - Sixth International Conference on Data Mining, ICDM 2006

PB - IEEE

ER -

Kriegel HP, Pryakhin A, Schubert M, Zimek A. COSMIC: Conceptually specified multi-instance clusters. In Proceedings - Sixth International Conference on Data Mining, ICDM 2006. IEEE. 2006. p. 917-921 https://doi.org/10.1109/ICDM.2006.46