Exceptionally monotone models : the rank correlation model class for Exceptional Model Mining

L. Downar, W. Duivesteijn

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

4 Citaties (Scopus)

Uittreksel

Exceptional Model Mining strives to find coherent subgroups of the dataset where multiple target attributes interact in an unusual way. One instance of such an investigated form of interaction is Pearson’s correlation coefficient between two targets. EMM then finds subgroups with an exceptionally linear relation between the targets. In this paper, we enrich the EMM toolbox by developing the more general rank correlation model class. We find subgroups with an exceptionally monotone relation between the targets. Apart from catering for this richer set of relations, the rank correlation model class does not necessarily require the assumption of target normality, which is implicitly invoked in the Pearson’s correlation model class. Furthermore, it is less sensitive to outliers. We provide pseudocode for the employed algorithm and analyze its computational complexity, and experimentally illustrate what the rank correlation model class for EMM can find for you on six datasets from an eclectic variety of domains.
Originele taal-2Engels
Pagina's (van-tot)369-394
TijdschriftKnowledge and Information Systems
Volume51
Nummer van het tijdschrift2
DOI's
StatusGepubliceerd - 2017
Extern gepubliceerdJa

Vingerafdruk

Computational complexity

Citeer dit

@article{cf6b08ff7b134143a31caadee63eab4a,
title = "Exceptionally monotone models : the rank correlation model class for Exceptional Model Mining",
abstract = "Exceptional Model Mining strives to find coherent subgroups of the dataset where multiple target attributes interact in an unusual way. One instance of such an investigated form of interaction is Pearson’s correlation coefficient between two targets. EMM then finds subgroups with an exceptionally linear relation between the targets. In this paper, we enrich the EMM toolbox by developing the more general rank correlation model class. We find subgroups with an exceptionally monotone relation between the targets. Apart from catering for this richer set of relations, the rank correlation model class does not necessarily require the assumption of target normality, which is implicitly invoked in the Pearson’s correlation model class. Furthermore, it is less sensitive to outliers. We provide pseudocode for the employed algorithm and analyze its computational complexity, and experimentally illustrate what the rank correlation model class for EMM can find for you on six datasets from an eclectic variety of domains.",
author = "L. Downar and W. Duivesteijn",
year = "2017",
doi = "10.1007/s10115-016-0979-z",
language = "English",
volume = "51",
pages = "369--394",
journal = "Knowledge and Information Systems",
issn = "0219-1377",
publisher = "Springer",
number = "2",

}

Exceptionally monotone models : the rank correlation model class for Exceptional Model Mining. / Downar, L.; Duivesteijn, W.

In: Knowledge and Information Systems, Vol. 51, Nr. 2, 2017, blz. 369-394.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Exceptionally monotone models : the rank correlation model class for Exceptional Model Mining

AU - Downar, L.

AU - Duivesteijn, W.

PY - 2017

Y1 - 2017

N2 - Exceptional Model Mining strives to find coherent subgroups of the dataset where multiple target attributes interact in an unusual way. One instance of such an investigated form of interaction is Pearson’s correlation coefficient between two targets. EMM then finds subgroups with an exceptionally linear relation between the targets. In this paper, we enrich the EMM toolbox by developing the more general rank correlation model class. We find subgroups with an exceptionally monotone relation between the targets. Apart from catering for this richer set of relations, the rank correlation model class does not necessarily require the assumption of target normality, which is implicitly invoked in the Pearson’s correlation model class. Furthermore, it is less sensitive to outliers. We provide pseudocode for the employed algorithm and analyze its computational complexity, and experimentally illustrate what the rank correlation model class for EMM can find for you on six datasets from an eclectic variety of domains.

AB - Exceptional Model Mining strives to find coherent subgroups of the dataset where multiple target attributes interact in an unusual way. One instance of such an investigated form of interaction is Pearson’s correlation coefficient between two targets. EMM then finds subgroups with an exceptionally linear relation between the targets. In this paper, we enrich the EMM toolbox by developing the more general rank correlation model class. We find subgroups with an exceptionally monotone relation between the targets. Apart from catering for this richer set of relations, the rank correlation model class does not necessarily require the assumption of target normality, which is implicitly invoked in the Pearson’s correlation model class. Furthermore, it is less sensitive to outliers. We provide pseudocode for the employed algorithm and analyze its computational complexity, and experimentally illustrate what the rank correlation model class for EMM can find for you on six datasets from an eclectic variety of domains.

U2 - 10.1007/s10115-016-0979-z

DO - 10.1007/s10115-016-0979-z

M3 - Article

VL - 51

SP - 369

EP - 394

JO - Knowledge and Information Systems

JF - Knowledge and Information Systems

SN - 0219-1377

IS - 2

ER -