On using Additional Unlabeled Data for Improving Dissimilarity-Based Classifications

Sang-Woon Kim

2013

Abstract

This paper reports an experimental result obtained with additionally using unlabeled data together with labeled ones to improve the classification accuracy of dissimilarity-based methods, namely, dissimilarity-based classifications (DBC) (Pe¸kalska, E. and Duin, R. P .W., 2005). In DBC, classifiers among classes are not based on the feature measurements of individual objects, but rather on a suitable dissimilarity measure among the objects. In image classification tasks, on the other hand, one of the intractable problems is the lack of information caused by the insufficient number of data. To address this problem in DBC, in this paper we study a new way of measuring the dissimilarity distance between two object images by using the well-known one-shot similarity metric (OSS) (Wolf, L. et al., 2009). In DBC using OSS, the dissimilarity distance is measured based on unlabeled (background) data that do not belong to the classes being learned, and consequently, do not require labeling. From this point of view, the classification is done in a semi-supervised learning (SSL) framework. Our experimental results, obtained with well-known benchmarks, demonstrate that when the cardinalities of the unlabeled data set and the prototype set have been appropriately chosen using additional unlabeled data for the OSS metric in SSL, DBC can be improved in terms of classification accuracies.

Paper Citation

in Harvard Style

Kim S. (2013). On using Additional Unlabeled Data for Improving Dissimilarity-Based Classifications . In Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-8565-41-9, pages 132-137. DOI: 10.5220/0004218901320137

in Bibtex Style

@conference{icpram13,
author={Sang-Woon Kim},
title={On using Additional Unlabeled Data for Improving Dissimilarity-Based Classifications},
booktitle={Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2013},
pages={132-137},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004218901320137},
isbn={978-989-8565-41-9},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - On using Additional Unlabeled Data for Improving Dissimilarity-Based Classifications
SN - 978-989-8565-41-9
AU - Kim S.
PY - 2013
SP - 132
EP - 137
DO - 10.5220/0004218901320137