Hierarchical feature selection for biological data

Witold Jacak; Karin Pröll

Hierarchical feature selection for biological data

Witold Jacak, Karin Pröll

Research Center Hagenberg

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

Abstract

In this paper we present feature selection in biological data by combining unsupervised learning with supervised cross validation. Unsupervised clustering methods are used to perform a clustering of object-data for a chosen subset of input features and given number of clusters. The resulting object clusters are compared with the predefined original object classes and a matching factor (score) is calculated. This score is used as criterion function for heuristic sequential feature selection and a cross selection algorithm.

Original language	English
Title of host publication	26th European Modeling and Simulation Symposium, EMSS 2014
Editors	Yuri Merkuryev, Lin Zhang, Emilio Jimenez, Francesco Longo, Michael Affenzeller, Agostino G. Bruzzone
Publisher	DIPTEM University of Genova
Pages	93-97
Number of pages	5
ISBN (Electronic)	9788897999324
ISBN (Print)	978-88-97999-38-6
Publication status	Published - 2014
Event	The 26th European Modeling & Simulation Symposium EMSS 2014 - Bordeaux, France Duration: 10 Sept 2014 → 12 Sept 2014 http://www.msc-les.org/conf/emss2014/index.htm

Publication series

Name	26th European Modeling and Simulation Symposium, EMSS 2014

Conference

Conference	The 26th European Modeling & Simulation Symposium EMSS 2014
Country/Territory	France
City	Bordeaux
Period	10.09.2014 → 12.09.2014
Internet address	http://www.msc-les.org/conf/emss2014/index.htm

Keywords

classification
clustering
feature selection
sequential feature selection
Sequential feature selection
Feature selection
Classification
Clustering

Cite this

@inproceedings{9bc3a91f9ea342268f89d8c09b869401,

title = "Hierarchical feature selection for biological data",

abstract = "In this paper we present feature selection in biological data by combining unsupervised learning with supervised cross validation. Unsupervised clustering methods are used to perform a clustering of object-data for a chosen subset of input features and given number of clusters. The resulting object clusters are compared with the predefined original object classes and a matching factor (score) is calculated. This score is used as criterion function for heuristic sequential feature selection and a cross selection algorithm.",

keywords = "classification, clustering, feature selection, sequential feature selection, classification, clustering, feature selection, sequential feature selection, Sequential feature selection, Feature selection, Classification, Clustering",

author = "Witold Jacak and Karin Pr{\"o}ll",

year = "2014",

language = "English",

isbn = "978-88-97999-38-6",

series = "26th European Modeling and Simulation Symposium, EMSS 2014",

publisher = "DIPTEM University of Genova",

pages = "93--97",

editor = "Yuri Merkuryev and Lin Zhang and Emilio Jimenez and Francesco Longo and Michael Affenzeller and Bruzzone, {Agostino G.}",

booktitle = "26th European Modeling and Simulation Symposium, EMSS 2014",

note = "The 26th European Modeling & Simulation Symposium EMSS 2014 ; Conference date: 10-09-2014 Through 12-09-2014",

url = "http://www.msc-les.org/conf/emss2014/index.htm",

}

Jacak, W & Pröll, K 2014, Hierarchical feature selection for biological data. in Y Merkuryev, L Zhang, E Jimenez, F Longo, M Affenzeller & AG Bruzzone (eds), 26th European Modeling and Simulation Symposium, EMSS 2014. 26th European Modeling and Simulation Symposium, EMSS 2014, DIPTEM University of Genova, pp. 93-97, The 26th European Modeling & Simulation Symposium EMSS 2014, Bordeaux, France, 10.09.2014.

Hierarchical feature selection for biological data. / Jacak, Witold; Pröll, Karin.
26th European Modeling and Simulation Symposium, EMSS 2014. ed. / Yuri Merkuryev; Lin Zhang; Emilio Jimenez; Francesco Longo; Michael Affenzeller; Agostino G. Bruzzone. DIPTEM University of Genova, 2014. p. 93-97 (26th European Modeling and Simulation Symposium, EMSS 2014).

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

TY - GEN

T1 - Hierarchical feature selection for biological data

AU - Jacak, Witold

AU - Pröll, Karin

PY - 2014

Y1 - 2014

N2 - In this paper we present feature selection in biological data by combining unsupervised learning with supervised cross validation. Unsupervised clustering methods are used to perform a clustering of object-data for a chosen subset of input features and given number of clusters. The resulting object clusters are compared with the predefined original object classes and a matching factor (score) is calculated. This score is used as criterion function for heuristic sequential feature selection and a cross selection algorithm.

AB - In this paper we present feature selection in biological data by combining unsupervised learning with supervised cross validation. Unsupervised clustering methods are used to perform a clustering of object-data for a chosen subset of input features and given number of clusters. The resulting object clusters are compared with the predefined original object classes and a matching factor (score) is calculated. This score is used as criterion function for heuristic sequential feature selection and a cross selection algorithm.

KW - classification

KW - clustering

KW - feature selection

KW - sequential feature selection

KW - classification

KW - clustering

KW - feature selection

KW - sequential feature selection

KW - Sequential feature selection

KW - Feature selection

KW - Classification

KW - Clustering

UR - http://www.scopus.com/inward/record.url?scp=84912099477&partnerID=8YFLogxK

M3 - Conference contribution

SN - 978-88-97999-38-6

T3 - 26th European Modeling and Simulation Symposium, EMSS 2014

SP - 93

EP - 97

BT - 26th European Modeling and Simulation Symposium, EMSS 2014

A2 - Merkuryev, Yuri

A2 - Zhang, Lin

A2 - Jimenez, Emilio

A2 - Longo, Francesco

A2 - Affenzeller, Michael

A2 - Bruzzone, Agostino G.

PB - DIPTEM University of Genova

T2 - The 26th European Modeling & Simulation Symposium EMSS 2014

Y2 - 10 September 2014 through 12 September 2014

ER -

Hierarchical feature selection for biological data

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this