Clustering Via Supervised Support Vector Machines

An SVM-based clustering algorithm is introduced that clusters data with no a priori knowledge of input classes. The algorithm initializes by first running a binary SVM classifier against a data set with each vector in the set randomly labeled. Once this initialization step is complete, the SVM co...

Full description

Bibliographic Details
Main Author:	Merat, Sepehr
Format:	Others
Published:	ScholarWorks@UNO 2008
Subjects:	clustering machine learning pattern recognition support vector machines supervised learning unsupervised learning
Online Access:	http://scholarworks.uno.edu/td/857 http://scholarworks.uno.edu/cgi/viewcontent.cgi?article=1837&context=td

id	ndltd-uno.edu-oai-scholarworks.uno.edu-td-1837
record_format	oai_dc
spelling	ndltd-uno.edu-oai-scholarworks.uno.edu-td-18372016-10-21T17:04:41Z Clustering Via Supervised Support Vector Machines Merat, Sepehr An SVM-based clustering algorithm is introduced that clusters data with no a priori knowledge of input classes. The algorithm initializes by first running a binary SVM classifier against a data set with each vector in the set randomly labeled. Once this initialization step is complete, the SVM confidence parameters for classification on each of the training instances can be accessed. The lowest confidence data (e.g., the worst of the mislabeled data) then has its labels switched to the other class label. The SVM is then re-run on the data set (with partly re-labeled data). The repetition of the above process improves the separability until there is no misclassification. Variations on this type of clustering approach are shown. 2008-08-07T07:00:00Z text application/pdf http://scholarworks.uno.edu/td/857 http://scholarworks.uno.edu/cgi/viewcontent.cgi?article=1837&context=td University of New Orleans Theses and Dissertations ScholarWorks@UNO clustering machine learning pattern recognition support vector machines supervised learning unsupervised learning
collection	NDLTD
format	Others
sources	NDLTD
topic	clustering machine learning pattern recognition support vector machines supervised learning unsupervised learning
spellingShingle	clustering machine learning pattern recognition support vector machines supervised learning unsupervised learning Merat, Sepehr Clustering Via Supervised Support Vector Machines
description	An SVM-based clustering algorithm is introduced that clusters data with no a priori knowledge of input classes. The algorithm initializes by first running a binary SVM classifier against a data set with each vector in the set randomly labeled. Once this initialization step is complete, the SVM confidence parameters for classification on each of the training instances can be accessed. The lowest confidence data (e.g., the worst of the mislabeled data) then has its labels switched to the other class label. The SVM is then re-run on the data set (with partly re-labeled data). The repetition of the above process improves the separability until there is no misclassification. Variations on this type of clustering approach are shown.
author	Merat, Sepehr
author_facet	Merat, Sepehr
author_sort	Merat, Sepehr
title	Clustering Via Supervised Support Vector Machines
title_short	Clustering Via Supervised Support Vector Machines
title_full	Clustering Via Supervised Support Vector Machines
title_fullStr	Clustering Via Supervised Support Vector Machines
title_full_unstemmed	Clustering Via Supervised Support Vector Machines
title_sort	clustering via supervised support vector machines
publisher	ScholarWorks@UNO
publishDate	2008
url	http://scholarworks.uno.edu/td/857 http://scholarworks.uno.edu/cgi/viewcontent.cgi?article=1837&context=td
work_keys_str_mv	AT meratsepehr clusteringviasupervisedsupportvectormachines
_version_	1718388016799547392

Clustering Via Supervised Support Vector Machines

Similar Items