"Kernelized" self-organizing maps for structured data

Fabio Aiolli, Giovanni Martino, Alessandro Sperduti, Markus Hagenbuchner

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

The suitability of the well known kernels for trees, and the lesser known Self- Organizing Map for Structures for categorization tasks on structured data is investigated in this paper. It is shown that a suitable combination of the two approaches, by defining new kernels on the activation map of a Self-Organizing Map for Structures, can result in a system that is significantly more accurate for categorization tasks on structured data. The effectiveness of the proposed approach is demonstrated experimentally on a relatively large corpus of XML formatted data.

Original languageEnglish
Title of host publicationESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks
Pages19-24
Number of pages6
Publication statusPublished - 2007
Externally publishedYes
Event15th European Symposium on Artificial Neural Networks, ESANN 2007 - Bruges
Duration: 25 Apr 200727 Apr 2007

Other

Other15th European Symposium on Artificial Neural Networks, ESANN 2007
CityBruges
Period25/4/0727/4/07

Fingerprint

Self organizing maps
XML
Chemical activation

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Cite this

Aiolli, F., Martino, G., Sperduti, A., & Hagenbuchner, M. (2007). "Kernelized" self-organizing maps for structured data. In ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks (pp. 19-24)

"Kernelized" self-organizing maps for structured data. / Aiolli, Fabio; Martino, Giovanni; Sperduti, Alessandro; Hagenbuchner, Markus.

ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks. 2007. p. 19-24.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Aiolli, F, Martino, G, Sperduti, A & Hagenbuchner, M 2007, "Kernelized" self-organizing maps for structured data. in ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks. pp. 19-24, 15th European Symposium on Artificial Neural Networks, ESANN 2007, Bruges, 25/4/07.
Aiolli F, Martino G, Sperduti A, Hagenbuchner M. "Kernelized" self-organizing maps for structured data. In ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks. 2007. p. 19-24
Aiolli, Fabio ; Martino, Giovanni ; Sperduti, Alessandro ; Hagenbuchner, Markus. / "Kernelized" self-organizing maps for structured data. ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks. 2007. pp. 19-24
@inproceedings{5c09007c90b040f3bbf095110e198577,
title = "{"}Kernelized{"} self-organizing maps for structured data",
abstract = "The suitability of the well known kernels for trees, and the lesser known Self- Organizing Map for Structures for categorization tasks on structured data is investigated in this paper. It is shown that a suitable combination of the two approaches, by defining new kernels on the activation map of a Self-Organizing Map for Structures, can result in a system that is significantly more accurate for categorization tasks on structured data. The effectiveness of the proposed approach is demonstrated experimentally on a relatively large corpus of XML formatted data.",
author = "Fabio Aiolli and Giovanni Martino and Alessandro Sperduti and Markus Hagenbuchner",
year = "2007",
language = "English",
isbn = "2930307099",
pages = "19--24",
booktitle = "ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks",

}

TY - GEN

T1 - "Kernelized" self-organizing maps for structured data

AU - Aiolli, Fabio

AU - Martino, Giovanni

AU - Sperduti, Alessandro

AU - Hagenbuchner, Markus

PY - 2007

Y1 - 2007

N2 - The suitability of the well known kernels for trees, and the lesser known Self- Organizing Map for Structures for categorization tasks on structured data is investigated in this paper. It is shown that a suitable combination of the two approaches, by defining new kernels on the activation map of a Self-Organizing Map for Structures, can result in a system that is significantly more accurate for categorization tasks on structured data. The effectiveness of the proposed approach is demonstrated experimentally on a relatively large corpus of XML formatted data.

AB - The suitability of the well known kernels for trees, and the lesser known Self- Organizing Map for Structures for categorization tasks on structured data is investigated in this paper. It is shown that a suitable combination of the two approaches, by defining new kernels on the activation map of a Self-Organizing Map for Structures, can result in a system that is significantly more accurate for categorization tasks on structured data. The effectiveness of the proposed approach is demonstrated experimentally on a relatively large corpus of XML formatted data.

UR - http://www.scopus.com/inward/record.url?scp=72149086038&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=72149086038&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:72149086038

SN - 2930307099

SN - 9782930307091

SP - 19

EP - 24

BT - ESANN 2007 Proceedings - 15th European Symposium on Artificial Neural Networks

ER -