Document classification using semantic networks with an adaptive similarity measure

Ginter, Filip; Pyysalo, Sampo; Salakoski, Tapio

by Filip Ginter, Sampo Pyysalo, Tapio Salakoski

Abstract:

We consider supervised document classification where a semantic network is used to augment document features with their hypernyms. A novel document representation is introduced in which the contribution of the hypernyms to document similarity is determined by semantic network edge weights. We argue that the optimal edge weights are not a static property of the semantic network, but should rather be adapted to the given classification task. To determine the optimal weights, we introduce an e\cient gradient descent method driven by the misclassifications of the k-nearest neighbor (kNN) classifier.

View PDF

Reference:

Document classification using semantic networks with an adaptive similarity measure (Filip Ginter, Sampo Pyysalo, Tapio Salakoski), In Amsterdam studies in the theory and history of linguistic science series, Citeseer, volume 292, 2007.

Bibtex Entry:

@article{Ginter2007,
abstract = {We consider supervised document classification where a semantic network is used to augment document features with their hypernyms. A novel document representation is introduced in which the contribution of the hypernyms to document similarity is determined by semantic network edge weights. We argue that the optimal edge weights are not a static property of the semantic network, but should rather be adapted to the given classification task. To determine the optimal weights, we introduce an e\#cient gradient descent method driven by the misclassifications of the k-nearest neighbor (kNN) classifier.},
author = {Ginter, Filip and Pyysalo, Sampo and Salakoski, Tapio},
journal = {Amsterdam studies in the theory and history of linguistic science series},
keywords = {SML-LIB-BIBLIO,lang:ENG},
mendeley-tags = {SML-LIB-BIBLIO,lang:ENG},
number = {4},
pages = {137},
publisher = {Citeseer},
title = {{Document classification using semantic networks with an adaptive similarity measure}},
url = {http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.60.3214\&rep=rep1\&type=pdf},
volume = {292},
year = {2007}
}