Modeling toxicity by using supervised Kohonen neural networks

Paolo Mazzatorta, Marjan Vračko, Aneta Jezierska, Emilio Benfenati

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

Counterprogation neural network is shown to be a powerful and suitable tool for the investigation of toxicity. This study mined a data set of 568 chemicals. Two hundred eighty-two objects were used as the training set and 286 as the test set. The final model developed presents high performances on the data set R2 = 0.83 (R2 = 0.97 on the training set, R2 = 0.59 on the test set). This technique distinguishes itself also for the ability to give to the expert two-dimensional maps suitable for the study of the distribution/clustering of the data and the identification of outliers.

Original languageEnglish
Pages (from-to)485-492
Number of pages8
JournalJournal of Chemical Information and Computer Sciences
Volume43
Issue number2
DOIs
Publication statusPublished - Mar 2003

Fingerprint

Toxicity
Neural networks

ASJC Scopus subject areas

  • Chemistry(all)
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Information Systems

Cite this

Modeling toxicity by using supervised Kohonen neural networks. / Mazzatorta, Paolo; Vračko, Marjan; Jezierska, Aneta; Benfenati, Emilio.

In: Journal of Chemical Information and Computer Sciences, Vol. 43, No. 2, 03.2003, p. 485-492.

Research output: Contribution to journalArticle

Mazzatorta, Paolo ; Vračko, Marjan ; Jezierska, Aneta ; Benfenati, Emilio. / Modeling toxicity by using supervised Kohonen neural networks. In: Journal of Chemical Information and Computer Sciences. 2003 ; Vol. 43, No. 2. pp. 485-492.
@article{91634f5120454403983a11d6f5bdcaba,
title = "Modeling toxicity by using supervised Kohonen neural networks",
abstract = "Counterprogation neural network is shown to be a powerful and suitable tool for the investigation of toxicity. This study mined a data set of 568 chemicals. Two hundred eighty-two objects were used as the training set and 286 as the test set. The final model developed presents high performances on the data set R2 = 0.83 (R2 = 0.97 on the training set, R2 = 0.59 on the test set). This technique distinguishes itself also for the ability to give to the expert two-dimensional maps suitable for the study of the distribution/clustering of the data and the identification of outliers.",
author = "Paolo Mazzatorta and Marjan Vračko and Aneta Jezierska and Emilio Benfenati",
year = "2003",
month = "3",
doi = "10.1021/ci0256182",
language = "English",
volume = "43",
pages = "485--492",
journal = "Journal of Chemical Information and Computer Sciences",
issn = "0095-2338",
publisher = "American Chemical Society",
number = "2",

}

TY - JOUR

T1 - Modeling toxicity by using supervised Kohonen neural networks

AU - Mazzatorta, Paolo

AU - Vračko, Marjan

AU - Jezierska, Aneta

AU - Benfenati, Emilio

PY - 2003/3

Y1 - 2003/3

N2 - Counterprogation neural network is shown to be a powerful and suitable tool for the investigation of toxicity. This study mined a data set of 568 chemicals. Two hundred eighty-two objects were used as the training set and 286 as the test set. The final model developed presents high performances on the data set R2 = 0.83 (R2 = 0.97 on the training set, R2 = 0.59 on the test set). This technique distinguishes itself also for the ability to give to the expert two-dimensional maps suitable for the study of the distribution/clustering of the data and the identification of outliers.

AB - Counterprogation neural network is shown to be a powerful and suitable tool for the investigation of toxicity. This study mined a data set of 568 chemicals. Two hundred eighty-two objects were used as the training set and 286 as the test set. The final model developed presents high performances on the data set R2 = 0.83 (R2 = 0.97 on the training set, R2 = 0.59 on the test set). This technique distinguishes itself also for the ability to give to the expert two-dimensional maps suitable for the study of the distribution/clustering of the data and the identification of outliers.

UR - http://www.scopus.com/inward/record.url?scp=0345381744&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0345381744&partnerID=8YFLogxK

U2 - 10.1021/ci0256182

DO - 10.1021/ci0256182

M3 - Article

C2 - 12653512

AN - SCOPUS:0345381744

VL - 43

SP - 485

EP - 492

JO - Journal of Chemical Information and Computer Sciences

JF - Journal of Chemical Information and Computer Sciences

SN - 0095-2338

IS - 2

ER -