Subjectively Interesting Subgroup Discovery on Real-valued Targets

Jefrey Lijffijt; Bo  Kang; Wouter Duivesteijn; Kai Puolamäki; Emilia Oikarinen; Tijl De Bie

doi:10.1109/ICDE.2018.00148

Subjectively Interesting Subgroup Discovery on Real-valued Targets

Jefrey Lijffijt, Bo Kang, Wouter Duivesteijn, Kai Puolamäki, Emilia Oikarinen, Tijl De Bie

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

2 Citations (Scopus)

Abstract

Deriving insights from high-dimensional data is one of the core problems in data mining. The difficulty mainly stems from the large number of variable combinations to potentially consider. Hence, an obvious question is whether we can automate the search for interesting patterns. Here, we consider the setting where a user wants to learn as efficiently as possible about real-valued attributes. We introduce a method to find subgroups in the data that are maximally informative (in the Information Theoretic sense) with respect to one or more real-valued target attributes. The succinct subgroup descriptions are in terms of arbitrarily-Typed description attributes. The approach is based on the Subjective Interestingness framework FORSIED to use prior knowledge when mining most informative patterns.

Original language	English
Title of host publication	Proceedings of the 34th IEEE International Conference on Data Engineering (ICDE 2018)
Publisher	IEEE
Pages	1356-1359
Number of pages	4
ISBN (Electronic)	9781538655207
DOIs	https://doi.org/10.1109/ICDE.2018.00148
Publication status	Published - 2018
MoE publication type	A4 Article in a conference publication
Event	International Conference on Data Engineering - Paris, France Duration: 16 Apr 2018 → 19 Apr 2018 Conference number: 34

Conference

Conference	International Conference on Data Engineering
Abbreviated title	ICDE
Country	France
City	Paris
Period	16/04/2018 → 19/04/2018

Keywords

Exceptional Model Mining
Exploratory Data Mining
Pattern Mining
Subgroup Discovery
Subjective Interestingness

Access to Document

10.1109/ICDE.2018.00148

Fingerprint Dive into the research topics of 'Subjectively Interesting Subgroup Discovery on Real-valued Targets'. Together they form a unique fingerprint.

Cite this

@inproceedings{c3a90478549d4d86833ef9c2aeef9551,

title = "Subjectively Interesting Subgroup Discovery on Real-valued Targets",

abstract = "Deriving insights from high-dimensional data is one of the core problems in data mining. The difficulty mainly stems from the large number of variable combinations to potentially consider. Hence, an obvious question is whether we can automate the search for interesting patterns. Here, we consider the setting where a user wants to learn as efficiently as possible about real-valued attributes. We introduce a method to find subgroups in the data that are maximally informative (in the Information Theoretic sense) with respect to one or more real-valued target attributes. The succinct subgroup descriptions are in terms of arbitrarily-Typed description attributes. The approach is based on the Subjective Interestingness framework FORSIED to use prior knowledge when mining most informative patterns.",

keywords = "Exceptional Model Mining, Exploratory Data Mining, Pattern Mining, Subgroup Discovery, Subjective Interestingness",

author = "Jefrey Lijffijt and Bo Kang and Wouter Duivesteijn and Kai Puolam{\"a}ki and Emilia Oikarinen and {De Bie}, Tijl",

note = "| openaire: EC/H2020/665501/EU//PEGASUS-2 ; International Conference on Data Engineering, ICDE ; Conference date: 16-04-2018 Through 19-04-2018",

year = "2018",

doi = "10.1109/ICDE.2018.00148",

language = "English",

pages = "1356--1359",

booktitle = "Proceedings of the 34th IEEE International Conference on Data Engineering (ICDE 2018)",

publisher = "IEEE",

address = "United States",

}

Lijffijt, J, Kang, B, Duivesteijn, W, Puolamäki, K, Oikarinen, E & De Bie, T 2018, Subjectively Interesting Subgroup Discovery on Real-valued Targets. in Proceedings of the 34th IEEE International Conference on Data Engineering (ICDE 2018)., 8509369, IEEE, pp. 1356-1359, International Conference on Data Engineering, Paris, France, 16/04/2018. https://doi.org/10.1109/ICDE.2018.00148

Subjectively Interesting Subgroup Discovery on Real-valued Targets. / Lijffijt, Jefrey; Kang, Bo ; Duivesteijn, Wouter; Puolamäki, Kai; Oikarinen, Emilia; De Bie, Tijl.

Proceedings of the 34th IEEE International Conference on Data Engineering (ICDE 2018). IEEE, 2018. p. 1356-1359 8509369.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Subjectively Interesting Subgroup Discovery on Real-valued Targets

AU - Lijffijt, Jefrey

AU - Kang, Bo

AU - Duivesteijn, Wouter

AU - Puolamäki, Kai

AU - Oikarinen, Emilia

AU - De Bie, Tijl

N1 - Conference code: 34

PY - 2018

Y1 - 2018

N2 - Deriving insights from high-dimensional data is one of the core problems in data mining. The difficulty mainly stems from the large number of variable combinations to potentially consider. Hence, an obvious question is whether we can automate the search for interesting patterns. Here, we consider the setting where a user wants to learn as efficiently as possible about real-valued attributes. We introduce a method to find subgroups in the data that are maximally informative (in the Information Theoretic sense) with respect to one or more real-valued target attributes. The succinct subgroup descriptions are in terms of arbitrarily-Typed description attributes. The approach is based on the Subjective Interestingness framework FORSIED to use prior knowledge when mining most informative patterns.

AB - Deriving insights from high-dimensional data is one of the core problems in data mining. The difficulty mainly stems from the large number of variable combinations to potentially consider. Hence, an obvious question is whether we can automate the search for interesting patterns. Here, we consider the setting where a user wants to learn as efficiently as possible about real-valued attributes. We introduce a method to find subgroups in the data that are maximally informative (in the Information Theoretic sense) with respect to one or more real-valued target attributes. The succinct subgroup descriptions are in terms of arbitrarily-Typed description attributes. The approach is based on the Subjective Interestingness framework FORSIED to use prior knowledge when mining most informative patterns.

KW - Exceptional Model Mining

KW - Exploratory Data Mining

KW - Pattern Mining

KW - Subgroup Discovery

KW - Subjective Interestingness

U2 - 10.1109/ICDE.2018.00148

DO - 10.1109/ICDE.2018.00148

M3 - Conference contribution

SP - 1356

EP - 1359

BT - Proceedings of the 34th IEEE International Conference on Data Engineering (ICDE 2018)

PB - IEEE

T2 - International Conference on Data Engineering

Y2 - 16 April 2018 through 19 April 2018

ER -