Integrating supervised and unsupervised learning in self organizing maps for gene expression data analysis

Seferina Mavroudi, Andrei Dragomir, Stergios Papadimitriou, Anastasios Bezerianos

Research output: Contribution to journalArticle

Abstract

Recently, Self Organizing Maps have been a popular approach to analyze gene expression data. Our paper presents an improved SOM-based algorithm called Supervised Network Self Organizing Map (sNet-SOM), which overcomes the main drawbacks of existing techniques by adaptively determining the number of clusters with a dynamic extension process and integrating unsupcrvised and supervised learning in an effort to make use of prior knowledge on data. The process is driven by an inhomogeneous measure that balances unsupervised/supervised learning and model complexity criteria. Multiple models are dynamically constructed by the algorithm, each corresponding to an unsupervised/supervised balance, model selection criteria being used to select the optimum one. The design allows us to effectively utilize multiple functional class labeling.

Original languageEnglish (US)
Pages (from-to)262-270
Number of pages9
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2714
StatePublished - 2003
Externally publishedYes

Fingerprint

Unsupervised learning
Unsupervised Learning
Supervised learning
Self organizing maps
Supervised Learning
Self-organizing Map
Gene Expression Data
Gene expression
Data analysis
Learning
Gene Expression
Model Selection Criteria
Model Complexity
Multiple Models
Number of Clusters
Prior Knowledge
Patient Selection
Labeling
Class
Design

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

Integrating supervised and unsupervised learning in self organizing maps for gene expression data analysis. / Mavroudi, Seferina; Dragomir, Andrei; Papadimitriou, Stergios; Bezerianos, Anastasios.

In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 2714, 2003, p. 262-270.

Research output: Contribution to journalArticle

@article{29530b37255c4cc4ac5fa10e28d7cdf4,
title = "Integrating supervised and unsupervised learning in self organizing maps for gene expression data analysis",
abstract = "Recently, Self Organizing Maps have been a popular approach to analyze gene expression data. Our paper presents an improved SOM-based algorithm called Supervised Network Self Organizing Map (sNet-SOM), which overcomes the main drawbacks of existing techniques by adaptively determining the number of clusters with a dynamic extension process and integrating unsupcrvised and supervised learning in an effort to make use of prior knowledge on data. The process is driven by an inhomogeneous measure that balances unsupervised/supervised learning and model complexity criteria. Multiple models are dynamically constructed by the algorithm, each corresponding to an unsupervised/supervised balance, model selection criteria being used to select the optimum one. The design allows us to effectively utilize multiple functional class labeling.",
author = "Seferina Mavroudi and Andrei Dragomir and Stergios Papadimitriou and Anastasios Bezerianos",
year = "2003",
language = "English (US)",
volume = "2714",
pages = "262--270",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Integrating supervised and unsupervised learning in self organizing maps for gene expression data analysis

AU - Mavroudi, Seferina

AU - Dragomir, Andrei

AU - Papadimitriou, Stergios

AU - Bezerianos, Anastasios

PY - 2003

Y1 - 2003

N2 - Recently, Self Organizing Maps have been a popular approach to analyze gene expression data. Our paper presents an improved SOM-based algorithm called Supervised Network Self Organizing Map (sNet-SOM), which overcomes the main drawbacks of existing techniques by adaptively determining the number of clusters with a dynamic extension process and integrating unsupcrvised and supervised learning in an effort to make use of prior knowledge on data. The process is driven by an inhomogeneous measure that balances unsupervised/supervised learning and model complexity criteria. Multiple models are dynamically constructed by the algorithm, each corresponding to an unsupervised/supervised balance, model selection criteria being used to select the optimum one. The design allows us to effectively utilize multiple functional class labeling.

AB - Recently, Self Organizing Maps have been a popular approach to analyze gene expression data. Our paper presents an improved SOM-based algorithm called Supervised Network Self Organizing Map (sNet-SOM), which overcomes the main drawbacks of existing techniques by adaptively determining the number of clusters with a dynamic extension process and integrating unsupcrvised and supervised learning in an effort to make use of prior knowledge on data. The process is driven by an inhomogeneous measure that balances unsupervised/supervised learning and model complexity criteria. Multiple models are dynamically constructed by the algorithm, each corresponding to an unsupervised/supervised balance, model selection criteria being used to select the optimum one. The design allows us to effectively utilize multiple functional class labeling.

UR - http://www.scopus.com/inward/record.url?scp=35248852352&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35248852352&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:35248852352

VL - 2714

SP - 262

EP - 270

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -