Discovering a taste for the unusual: exceptional models for preference mining

Cláudio Rebelo de Sá, Wouter Duivesteijn, Paulo Azevedo, Alípio Mário Jorge, Carlos Soares, Arno Knobbe

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

2 Citaties (Scopus)
27 Downloads (Pure)

Uittreksel

Exceptional preferences mining (EPM) is a crossover between two subfields of data mining: local pattern mining and preference learning. EPM can be seen as a local pattern mining task that finds subsets of observations where some preference relations between labels significantly deviate from the norm. It is a variant of subgroup discovery, with rankings of labels as the target concept. We employ several quality measures that highlight subgroups featuring exceptional preferences, where the focus of what constitutes ‘exceptional’ varies with the quality measure: two measures look for exceptional overall ranking behavior, one measure indicates whether a particular label stands out from the rest, and a fourth measure highlights subgroups with unusual pairwise label ranking behavior. We explore a few datasets and compare with existing techniques. The results confirm that the new task EPM can deliver interesting knowledge.

Originele taal-2Engels
Pagina's (van-tot)1775-1807
Aantal pagina's33
TijdschriftMachine Learning
Volume107
Nummer van het tijdschrift11
DOI's
StatusGepubliceerd - 1 nov 2018

Vingerafdruk

Labels
Data mining

Citeer dit

de Sá, Cláudio Rebelo ; Duivesteijn, Wouter ; Azevedo, Paulo ; Jorge, Alípio Mário ; Soares, Carlos ; Knobbe, Arno. / Discovering a taste for the unusual : exceptional models for preference mining. In: Machine Learning. 2018 ; Vol. 107, Nr. 11. blz. 1775-1807.
@article{b552791beab04b73bf9b5c3451e0e5f4,
title = "Discovering a taste for the unusual: exceptional models for preference mining",
abstract = "Exceptional preferences mining (EPM) is a crossover between two subfields of data mining: local pattern mining and preference learning. EPM can be seen as a local pattern mining task that finds subsets of observations where some preference relations between labels significantly deviate from the norm. It is a variant of subgroup discovery, with rankings of labels as the target concept. We employ several quality measures that highlight subgroups featuring exceptional preferences, where the focus of what constitutes ‘exceptional’ varies with the quality measure: two measures look for exceptional overall ranking behavior, one measure indicates whether a particular label stands out from the rest, and a fourth measure highlights subgroups with unusual pairwise label ranking behavior. We explore a few datasets and compare with existing techniques. The results confirm that the new task EPM can deliver interesting knowledge.",
keywords = "Distribution rules, Exceptional model mining, Label ranking, Preference learning, Subgroup discovery",
author = "{de S{\'a}}, {Cl{\'a}udio Rebelo} and Wouter Duivesteijn and Paulo Azevedo and Jorge, {Al{\'i}pio M{\'a}rio} and Carlos Soares and Arno Knobbe",
year = "2018",
month = "11",
day = "1",
doi = "10.1007/s10994-018-5743-z",
language = "English",
volume = "107",
pages = "1775--1807",
journal = "Machine Learning",
issn = "0885-6125",
publisher = "Springer",
number = "11",

}

de Sá, CR, Duivesteijn, W, Azevedo, P, Jorge, AM, Soares, C & Knobbe, A 2018, 'Discovering a taste for the unusual: exceptional models for preference mining', Machine Learning, vol. 107, nr. 11, blz. 1775-1807. https://doi.org/10.1007/s10994-018-5743-z

Discovering a taste for the unusual : exceptional models for preference mining. / de Sá, Cláudio Rebelo; Duivesteijn, Wouter; Azevedo, Paulo; Jorge, Alípio Mário; Soares, Carlos; Knobbe, Arno.

In: Machine Learning, Vol. 107, Nr. 11, 01.11.2018, blz. 1775-1807.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Discovering a taste for the unusual

T2 - exceptional models for preference mining

AU - de Sá, Cláudio Rebelo

AU - Duivesteijn, Wouter

AU - Azevedo, Paulo

AU - Jorge, Alípio Mário

AU - Soares, Carlos

AU - Knobbe, Arno

PY - 2018/11/1

Y1 - 2018/11/1

N2 - Exceptional preferences mining (EPM) is a crossover between two subfields of data mining: local pattern mining and preference learning. EPM can be seen as a local pattern mining task that finds subsets of observations where some preference relations between labels significantly deviate from the norm. It is a variant of subgroup discovery, with rankings of labels as the target concept. We employ several quality measures that highlight subgroups featuring exceptional preferences, where the focus of what constitutes ‘exceptional’ varies with the quality measure: two measures look for exceptional overall ranking behavior, one measure indicates whether a particular label stands out from the rest, and a fourth measure highlights subgroups with unusual pairwise label ranking behavior. We explore a few datasets and compare with existing techniques. The results confirm that the new task EPM can deliver interesting knowledge.

AB - Exceptional preferences mining (EPM) is a crossover between two subfields of data mining: local pattern mining and preference learning. EPM can be seen as a local pattern mining task that finds subsets of observations where some preference relations between labels significantly deviate from the norm. It is a variant of subgroup discovery, with rankings of labels as the target concept. We employ several quality measures that highlight subgroups featuring exceptional preferences, where the focus of what constitutes ‘exceptional’ varies with the quality measure: two measures look for exceptional overall ranking behavior, one measure indicates whether a particular label stands out from the rest, and a fourth measure highlights subgroups with unusual pairwise label ranking behavior. We explore a few datasets and compare with existing techniques. The results confirm that the new task EPM can deliver interesting knowledge.

KW - Distribution rules

KW - Exceptional model mining

KW - Label ranking

KW - Preference learning

KW - Subgroup discovery

UR - http://www.scopus.com/inward/record.url?scp=85049622722&partnerID=8YFLogxK

U2 - 10.1007/s10994-018-5743-z

DO - 10.1007/s10994-018-5743-z

M3 - Article

AN - SCOPUS:85049622722

VL - 107

SP - 1775

EP - 1807

JO - Machine Learning

JF - Machine Learning

SN - 0885-6125

IS - 11

ER -