Expression inference - genetic symbolic classification integrated with non-linear coefficient optimisation

Hunter, Andrew (2002) Expression inference - genetic symbolic classification integrated with non-linear coefficient optimisation. In: Joint International Conferences, AISC 2002 & Calculemus 2002, 1-5 July 2002, Marseilles, France.

Documents
Expression inference - genetic symbolic classification integrated with non-linear coefficient optimisation
This paper introduces a general framework for inferring symbolic classifiers, using the Genetic Programming paradism with non-linear optimisation of embedded co-efficients.
[img]
[Download]
Request a copy
[img] PDF
ExpressionInference.pdf
Restricted to Registered users only

187kB

Official URL: http://www.informatik.uni-trier.de/~ley/db/conf/ai...

Abstract

Abstract. Expression Inference is a parsimonious, comprehensible alternative to semi-parametric and non-parametric classification techniques such as neural networks, which generates compact symbolic mathematical
expressions for classification or regression. This paper introduces a general framework for inferring symbolic classifiers, using the Genetic Programming paradigm with non-linear optimisation of embedded coefficients.
An error propagation algorithm is introduced to support the
optimisation. A multiobjective variant of Genetic Programming provides a range of models trading off parsimony and classification performance, the latter measured by ROC curve analysis. The technique is shown to
develop extremely concise and effective models on a sample real-world problem domain.
Keywords. Symbolic Regression; Classification; Genetic Programming; ROC Curves; Multiobjective Optimisation.
Topic. Symbolic Computations for Expert Systems and Machine Learning.

Item Type:Conference or Workshop Item (Paper)
Additional Information:Abstract. Expression Inference is a parsimonious, comprehensible alternative to semi-parametric and non-parametric classification techniques such as neural networks, which generates compact symbolic mathematical expressions for classification or regression. This paper introduces a general framework for inferring symbolic classifiers, using the Genetic Programming paradigm with non-linear optimisation of embedded coefficients. An error propagation algorithm is introduced to support the optimisation. A multiobjective variant of Genetic Programming provides a range of models trading off parsimony and classification performance, the latter measured by ROC curve analysis. The technique is shown to develop extremely concise and effective models on a sample real-world problem domain. Keywords. Symbolic Regression; Classification; Genetic Programming; ROC Curves; Multiobjective Optimisation. Topic. Symbolic Computations for Expert Systems and Machine Learning.
Keywords:Symbolic regression, Classification, Genetic Programming, ROC Curves, Multiobjective optimisation
Subjects:G Mathematical and Computer Sciences > G760 Machine Learning
Divisions:College of Science > School of Computer Science
ID Code:1897
Deposited By: Tammie Farley
Deposited On:30 Jun 2009 15:23
Last Modified:13 Mar 2013 08:32

Repository Staff Only: item control page