Feature-Based Learners for Description Logics

Daniel Kudenko; Haym Hirsh

Feature-Based Learners for Description Logics

Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

Inductive learning algorithms that have been applied to learning in description logics (DL) have not been as well studied and optimized as the more general class of feature-based learning algorithms. This paper proposes a way to apply feature-based learners to DL learning tasks by presenting a method to compute a feature-vector representation for DL instances. The representation is based on concepts computed by a DL learning algorithm and by a feature generation method that has previously been applied to sequence categorization tasks. We show encouraging empirical test results using the feature-based learning systems Ripper, C5.0, and Naive Bayes. 1 Introduction Description logics (DL) are a well-studied formalism for the representation of knowledge, for which inductive learning problems have been defined in the past [3, 5]. When presented with a set of labeled examples (i.e., DL individuals), the learning algorithms compute a hypothesis to predict the label of new, previously unseen...

Original language	Undefined/Unknown
Publication status	Published - 1999

Access to Document

Cite this

@conference{ea6cf4e9645749be904b7cdd766096fb,

title = "Feature-Based Learners for Description Logics",

abstract = "Inductive learning algorithms that have been applied to learning in description logics (DL) have not been as well studied and optimized as the more general class of feature-based learning algorithms. This paper proposes a way to apply feature-based learners to DL learning tasks by presenting a method to compute a feature-vector representation for DL instances. The representation is based on concepts computed by a DL learning algorithm and by a feature generation method that has previously been applied to sequence categorization tasks. We show encouraging empirical test results using the feature-based learning systems Ripper, C5.0, and Naive Bayes. 1 Introduction Description logics (DL) are a well-studied formalism for the representation of knowledge, for which inductive learning problems have been defined in the past [3, 5]. When presented with a set of labeled examples (i.e., DL individuals), the learning algorithms compute a hypothesis to predict the label of new, previously unseen...",

author = "Daniel Kudenko and Haym Hirsh",

year = "1999",

language = "Undefined/Unknown",

}

TY - CONF

T1 - Feature-Based Learners for Description Logics

AU - Kudenko, Daniel

AU - Hirsh, Haym

PY - 1999

Y1 - 1999

N2 - Inductive learning algorithms that have been applied to learning in description logics (DL) have not been as well studied and optimized as the more general class of feature-based learning algorithms. This paper proposes a way to apply feature-based learners to DL learning tasks by presenting a method to compute a feature-vector representation for DL instances. The representation is based on concepts computed by a DL learning algorithm and by a feature generation method that has previously been applied to sequence categorization tasks. We show encouraging empirical test results using the feature-based learning systems Ripper, C5.0, and Naive Bayes. 1 Introduction Description logics (DL) are a well-studied formalism for the representation of knowledge, for which inductive learning problems have been defined in the past [3, 5]. When presented with a set of labeled examples (i.e., DL individuals), the learning algorithms compute a hypothesis to predict the label of new, previously unseen...

AB - Inductive learning algorithms that have been applied to learning in description logics (DL) have not been as well studied and optimized as the more general class of feature-based learning algorithms. This paper proposes a way to apply feature-based learners to DL learning tasks by presenting a method to compute a feature-vector representation for DL instances. The representation is based on concepts computed by a DL learning algorithm and by a feature generation method that has previously been applied to sequence categorization tasks. We show encouraging empirical test results using the feature-based learning systems Ripper, C5.0, and Naive Bayes. 1 Introduction Description logics (DL) are a well-studied formalism for the representation of knowledge, for which inductive learning problems have been defined in the past [3, 5]. When presented with a set of labeled examples (i.e., DL individuals), the learning algorithms compute a hypothesis to predict the label of new, previously unseen...

M3 - Paper

ER -