An Effective Combination of Multiple Classifiers for Toxicity Prediction

Gongde Guo, Daniel Neagu, Xuming Huang, Yaxin Bi

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

The performance of individual classifiers applied to complex data sets has for predictive toxicology a significant importance. An investigation was conducted to improve classification performance of combinations of classifiers. For this purpose some representative classification methods for individual classifier development have been used to assure a good range for model diversity. The paper proposes a new effective multi-classifier system based on Dempster’s rule of combination of individual classifiers. The performance of the new method has been evaluated on seven toxicity data sets. The classification accuracy of the proposed combination models achieved, according to our initial experiments, 2.97% better average than that of the best individual classifier among five classification methods (Instance-based Learning algorithm, Decision Tree, Repeated Incremental Pruning to Produce Error Reduction, Multi-Layer Perceptrons and Support Vector Machine) studied.
LanguageEnglish
Title of host publicationFuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science
Pages481-490
Publication statusPublished - 2006

Fingerprint

Toxicity
Classifiers
Multilayer neural networks
Decision trees
Learning algorithms
Support vector machines
Experiments

Keywords

  • Classification
  • Multiple Classifiers
  • Toxicity Prediction

Cite this

Guo, G., Neagu, D., Huang, X., & Bi, Y. (2006). An Effective Combination of Multiple Classifiers for Toxicity Prediction. In Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science (pp. 481-490)
Guo, Gongde ; Neagu, Daniel ; Huang, Xuming ; Bi, Yaxin. / An Effective Combination of Multiple Classifiers for Toxicity Prediction. Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science. 2006. pp. 481-490
@inbook{fd13d9c14be84979801428f013227c73,
title = "An Effective Combination of Multiple Classifiers for Toxicity Prediction",
abstract = "The performance of individual classifiers applied to complex data sets has for predictive toxicology a significant importance. An investigation was conducted to improve classification performance of combinations of classifiers. For this purpose some representative classification methods for individual classifier development have been used to assure a good range for model diversity. The paper proposes a new effective multi-classifier system based on Dempster’s rule of combination of individual classifiers. The performance of the new method has been evaluated on seven toxicity data sets. The classification accuracy of the proposed combination models achieved, according to our initial experiments, 2.97{\%} better average than that of the best individual classifier among five classification methods (Instance-based Learning algorithm, Decision Tree, Repeated Incremental Pruning to Produce Error Reduction, Multi-Layer Perceptrons and Support Vector Machine) studied.",
keywords = "Classification, Multiple Classifiers, Toxicity Prediction",
author = "Gongde Guo and Daniel Neagu and Xuming Huang and Yaxin Bi",
year = "2006",
language = "English",
isbn = "978-3-540-45916-3",
pages = "481--490",
booktitle = "Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science",

}

Guo, G, Neagu, D, Huang, X & Bi, Y 2006, An Effective Combination of Multiple Classifiers for Toxicity Prediction. in Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science. pp. 481-490.

An Effective Combination of Multiple Classifiers for Toxicity Prediction. / Guo, Gongde; Neagu, Daniel; Huang, Xuming; Bi, Yaxin.

Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science. 2006. p. 481-490.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - An Effective Combination of Multiple Classifiers for Toxicity Prediction

AU - Guo, Gongde

AU - Neagu, Daniel

AU - Huang, Xuming

AU - Bi, Yaxin

PY - 2006

Y1 - 2006

N2 - The performance of individual classifiers applied to complex data sets has for predictive toxicology a significant importance. An investigation was conducted to improve classification performance of combinations of classifiers. For this purpose some representative classification methods for individual classifier development have been used to assure a good range for model diversity. The paper proposes a new effective multi-classifier system based on Dempster’s rule of combination of individual classifiers. The performance of the new method has been evaluated on seven toxicity data sets. The classification accuracy of the proposed combination models achieved, according to our initial experiments, 2.97% better average than that of the best individual classifier among five classification methods (Instance-based Learning algorithm, Decision Tree, Repeated Incremental Pruning to Produce Error Reduction, Multi-Layer Perceptrons and Support Vector Machine) studied.

AB - The performance of individual classifiers applied to complex data sets has for predictive toxicology a significant importance. An investigation was conducted to improve classification performance of combinations of classifiers. For this purpose some representative classification methods for individual classifier development have been used to assure a good range for model diversity. The paper proposes a new effective multi-classifier system based on Dempster’s rule of combination of individual classifiers. The performance of the new method has been evaluated on seven toxicity data sets. The classification accuracy of the proposed combination models achieved, according to our initial experiments, 2.97% better average than that of the best individual classifier among five classification methods (Instance-based Learning algorithm, Decision Tree, Repeated Incremental Pruning to Produce Error Reduction, Multi-Layer Perceptrons and Support Vector Machine) studied.

KW - Classification

KW - Multiple Classifiers

KW - Toxicity Prediction

M3 - Chapter

SN - 978-3-540-45916-3

SP - 481

EP - 490

BT - Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science

ER -

Guo G, Neagu D, Huang X, Bi Y. An Effective Combination of Multiple Classifiers for Toxicity Prediction. In Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science. 2006. p. 481-490