Dissimilarity in Graph-Based Semi-Supervised Classification

Andrew B. Goldberg, Xiaojin Zhu, Stephen Wright
; Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, PMLR 2:155-162, 2007.

Abstract

Label dissimilarity specifies that a pair of examples probably have different class labels. We present a semi-supervised classification algorithm that learns from dissimilarity and similarity information on labeled and unlabeled data. Our approach uses a novel graphbased encoding of dissimilarity that results in a convex problem, and can handle both binary and multiclass classification. Experiments on several tasks are promising.

Cite this Paper


BibTeX
@InProceedings{pmlr-v2-goldberg07a, title = {Dissimilarity in Graph-Based Semi-Supervised Classification}, author = {Andrew B. Goldberg and Xiaojin Zhu and Stephen Wright}, booktitle = {Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics}, pages = {155--162}, year = {2007}, editor = {Marina Meila and Xiaotong Shen}, volume = {2}, series = {Proceedings of Machine Learning Research}, address = {San Juan, Puerto Rico}, month = {21--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v2/goldberg07a/goldberg07a.pdf}, url = {http://proceedings.mlr.press/v2/goldberg07a.html}, abstract = {Label dissimilarity specifies that a pair of examples probably have different class labels. We present a semi-supervised classification algorithm that learns from dissimilarity and similarity information on labeled and unlabeled data. Our approach uses a novel graphbased encoding of dissimilarity that results in a convex problem, and can handle both binary and multiclass classification. Experiments on several tasks are promising.} }
Endnote
%0 Conference Paper %T Dissimilarity in Graph-Based Semi-Supervised Classification %A Andrew B. Goldberg %A Xiaojin Zhu %A Stephen Wright %B Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2007 %E Marina Meila %E Xiaotong Shen %F pmlr-v2-goldberg07a %I PMLR %J Proceedings of Machine Learning Research %P 155--162 %U http://proceedings.mlr.press %V 2 %W PMLR %X Label dissimilarity specifies that a pair of examples probably have different class labels. We present a semi-supervised classification algorithm that learns from dissimilarity and similarity information on labeled and unlabeled data. Our approach uses a novel graphbased encoding of dissimilarity that results in a convex problem, and can handle both binary and multiclass classification. Experiments on several tasks are promising.
RIS
TY - CPAPER TI - Dissimilarity in Graph-Based Semi-Supervised Classification AU - Andrew B. Goldberg AU - Xiaojin Zhu AU - Stephen Wright BT - Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics PY - 2007/03/11 DA - 2007/03/11 ED - Marina Meila ED - Xiaotong Shen ID - pmlr-v2-goldberg07a PB - PMLR SP - 155 DP - PMLR EP - 162 L1 - http://proceedings.mlr.press/v2/goldberg07a/goldberg07a.pdf UR - http://proceedings.mlr.press/v2/goldberg07a.html AB - Label dissimilarity specifies that a pair of examples probably have different class labels. We present a semi-supervised classification algorithm that learns from dissimilarity and similarity information on labeled and unlabeled data. Our approach uses a novel graphbased encoding of dissimilarity that results in a convex problem, and can handle both binary and multiclass classification. Experiments on several tasks are promising. ER -
APA
Goldberg, A.B., Zhu, X. & Wright, S.. (2007). Dissimilarity in Graph-Based Semi-Supervised Classification. Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, in PMLR 2:155-162

Related Material