Unsupervised Neural Networks based Scoring and Feature Selection in Biological Data Analysis

Witold Jacak; Karin Pröll

Unsupervised Neural Networks based Scoring and Feature Selection in Biological Data Analysis

Witold Jacak, Karin Pröll

Research Center Hagenberg

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

Abstract

In this paper we present a novel method for scoring function specification and feature selection by combining unsupervised learning with supervised cross validation. A one dimensional Kohonen SOM is 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.

Original language	English
Title of host publication	Proceedings of IEEE APCAST'12 Conference
Pages	18-23
Publication status	Published - 2012
Event	IEEE APCast'12 - Sydney, Australia Duration: 6 Feb 2012 → 8 Feb 2012

Conference

Conference	IEEE APCast'12
Country/Territory	Australia
City	Sydney
Period	06.02.2012 → 08.02.2012

Keywords

classification
clustering
feature selection
Kohonen SOM
MLP

Cite this

@inproceedings{3823520880064c658441b2a145443a31,

title = "Unsupervised Neural Networks based Scoring and Feature Selection in Biological Data Analysis",

abstract = "In this paper we present a novel method for scoring function specification and feature selection by combining unsupervised learning with supervised cross validation. A one dimensional Kohonen SOM is 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.",

keywords = "classification, clustering, feature selection, Kohonen SOM, MLP, classification, clustering, feature selection, Kohonen SOM, MLP",

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

year = "2012",

language = "English",

pages = "18--23",

booktitle = "Proceedings of IEEE APCAST'12 Conference",

note = "IEEE APCast'12 ; Conference date: 06-02-2012 Through 08-02-2012",

}

TY - GEN

T1 - Unsupervised Neural Networks based Scoring and Feature Selection in Biological Data Analysis

AU - Jacak, Witold

AU - Pröll, Karin

PY - 2012

Y1 - 2012

N2 - In this paper we present a novel method for scoring function specification and feature selection by combining unsupervised learning with supervised cross validation. A one dimensional Kohonen SOM is 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.

AB - In this paper we present a novel method for scoring function specification and feature selection by combining unsupervised learning with supervised cross validation. A one dimensional Kohonen SOM is 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.

KW - classification

KW - clustering

KW - feature selection

KW - Kohonen SOM

KW - MLP

KW - classification

KW - clustering

KW - feature selection

KW - Kohonen SOM

KW - MLP

M3 - Conference contribution

SP - 18

EP - 23

BT - Proceedings of IEEE APCAST'12 Conference

T2 - IEEE APCast'12

Y2 - 6 February 2012 through 8 February 2012

ER -

Unsupervised Neural Networks based Scoring and Feature Selection in Biological Data Analysis

Abstract

Conference

Keywords

Fingerprint

Cite this