Multiclass classifier from a combination of local experts

Toward distributed computation for real-problem classifiers

Christoph König, Giuseppina Gini, Marian Craciun, Emilio Benfenati

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In many real-world applications simple classifiers are too weak to have predictive power. Ensemble techniques, or mixture of experts, are a possible solution. We illustrate why mixture of experts are a natural choice in domains such as the prediction of environmental toxicity for chemicals, when a structural approach is pursued. The real data here used are derived from peer reviewed experiments, and are publicly available, but are difficult to model. We used them to predict aquatic toxicity for fish. Chemical information was coded into a set of about 160 descriptors; after reducing the dimensions of the feature vector through different techniques, we developed multivariate regression to build a model of the toxic effects of chemicals. Defining toxicity as a category, as in European Union (EU) regulations, we extended the study to predict toxicity class. Problems with poor predictive power of this simple approach have led us to reconsider the problem from a more theoretical angle. We have respected locality criterion to build different local classifiers, one for each chemical class, to achieve better results. Then we combined the classifiers to get a complete system to predict any chemical for the chemical classes studied.

Original languageEnglish
Pages (from-to)801-817
Number of pages17
JournalInternational Journal of Pattern Recognition and Artificial Intelligence
Volume18
Issue number5
DOIs
Publication statusPublished - Aug 2004

Fingerprint

Classifiers
Toxicity
Fish
Experiments

Keywords

  • Classification from regression
  • Mixture of experts
  • QSAR

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition

Cite this

Multiclass classifier from a combination of local experts : Toward distributed computation for real-problem classifiers. / König, Christoph; Gini, Giuseppina; Craciun, Marian; Benfenati, Emilio.

In: International Journal of Pattern Recognition and Artificial Intelligence, Vol. 18, No. 5, 08.2004, p. 801-817.

Research output: Contribution to journalArticle

@article{528f4bf960bd4ccba204438c2dd795bb,
title = "Multiclass classifier from a combination of local experts: Toward distributed computation for real-problem classifiers",
abstract = "In many real-world applications simple classifiers are too weak to have predictive power. Ensemble techniques, or mixture of experts, are a possible solution. We illustrate why mixture of experts are a natural choice in domains such as the prediction of environmental toxicity for chemicals, when a structural approach is pursued. The real data here used are derived from peer reviewed experiments, and are publicly available, but are difficult to model. We used them to predict aquatic toxicity for fish. Chemical information was coded into a set of about 160 descriptors; after reducing the dimensions of the feature vector through different techniques, we developed multivariate regression to build a model of the toxic effects of chemicals. Defining toxicity as a category, as in European Union (EU) regulations, we extended the study to predict toxicity class. Problems with poor predictive power of this simple approach have led us to reconsider the problem from a more theoretical angle. We have respected locality criterion to build different local classifiers, one for each chemical class, to achieve better results. Then we combined the classifiers to get a complete system to predict any chemical for the chemical classes studied.",
keywords = "Classification from regression, Mixture of experts, QSAR",
author = "Christoph K{\"o}nig and Giuseppina Gini and Marian Craciun and Emilio Benfenati",
year = "2004",
month = "8",
doi = "10.1142/S0218001404003423",
language = "English",
volume = "18",
pages = "801--817",
journal = "International Journal of Pattern Recognition and Artificial Intelligence",
issn = "0218-0014",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "5",

}

TY - JOUR

T1 - Multiclass classifier from a combination of local experts

T2 - Toward distributed computation for real-problem classifiers

AU - König, Christoph

AU - Gini, Giuseppina

AU - Craciun, Marian

AU - Benfenati, Emilio

PY - 2004/8

Y1 - 2004/8

N2 - In many real-world applications simple classifiers are too weak to have predictive power. Ensemble techniques, or mixture of experts, are a possible solution. We illustrate why mixture of experts are a natural choice in domains such as the prediction of environmental toxicity for chemicals, when a structural approach is pursued. The real data here used are derived from peer reviewed experiments, and are publicly available, but are difficult to model. We used them to predict aquatic toxicity for fish. Chemical information was coded into a set of about 160 descriptors; after reducing the dimensions of the feature vector through different techniques, we developed multivariate regression to build a model of the toxic effects of chemicals. Defining toxicity as a category, as in European Union (EU) regulations, we extended the study to predict toxicity class. Problems with poor predictive power of this simple approach have led us to reconsider the problem from a more theoretical angle. We have respected locality criterion to build different local classifiers, one for each chemical class, to achieve better results. Then we combined the classifiers to get a complete system to predict any chemical for the chemical classes studied.

AB - In many real-world applications simple classifiers are too weak to have predictive power. Ensemble techniques, or mixture of experts, are a possible solution. We illustrate why mixture of experts are a natural choice in domains such as the prediction of environmental toxicity for chemicals, when a structural approach is pursued. The real data here used are derived from peer reviewed experiments, and are publicly available, but are difficult to model. We used them to predict aquatic toxicity for fish. Chemical information was coded into a set of about 160 descriptors; after reducing the dimensions of the feature vector through different techniques, we developed multivariate regression to build a model of the toxic effects of chemicals. Defining toxicity as a category, as in European Union (EU) regulations, we extended the study to predict toxicity class. Problems with poor predictive power of this simple approach have led us to reconsider the problem from a more theoretical angle. We have respected locality criterion to build different local classifiers, one for each chemical class, to achieve better results. Then we combined the classifiers to get a complete system to predict any chemical for the chemical classes studied.

KW - Classification from regression

KW - Mixture of experts

KW - QSAR

UR - http://www.scopus.com/inward/record.url?scp=8644279686&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=8644279686&partnerID=8YFLogxK

U2 - 10.1142/S0218001404003423

DO - 10.1142/S0218001404003423

M3 - Article

VL - 18

SP - 801

EP - 817

JO - International Journal of Pattern Recognition and Artificial Intelligence

JF - International Journal of Pattern Recognition and Artificial Intelligence

SN - 0218-0014

IS - 5

ER -