File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Building a Decision Cluster Classification Model for High Dimensional Data by a Variable Weighting k-Means Method

TitleBuilding a Decision Cluster Classification Model for High Dimensional Data by a Variable Weighting k-Means Method
Authors
Issue Date2008
PublisherSpringer-Verlag Berlin.
Citation
The 21st Australasian Joint Conference On Artificial Intelligence: Advances in Artificial Intelligence, Auckland, New Zealand, 1-5 December 2008, p. 337-347 How to Cite?
AbstractIn this paper, a new classification method (ADCC) for high dimensional data is proposed. In this method, a decision cluster classification model (DCC) consists of a set of disjoint decision clusters, each labeled with a dominant class that determines the class of new objects falling in the cluster. A cluster tree is first generated from a training data set by recursively calling a variable weighting k -means algorithm. Then, the DCC model is selected from the tree. Anderson-Darling test is used to determine the stopping condition of the tree growing. A series of experiments on both synthetic and real data sets have shown that the new classification method (ADCC) performed better in accuracy and scalability than the existing methods of k -NN , decision tree and SVM. It is particularly suitable for large, high dimensional data with many classes.
Persistent Identifierhttp://hdl.handle.net/10722/223760
ISBN

 

DC FieldValueLanguage
dc.contributor.authorLi, Y-
dc.contributor.authorHung, E-
dc.contributor.authorChung, K-
dc.contributor.authorHuang, JZ-
dc.date.accessioned2016-03-14T08:05:18Z-
dc.date.available2016-03-14T08:05:18Z-
dc.date.issued2008-
dc.identifier.citationThe 21st Australasian Joint Conference On Artificial Intelligence: Advances in Artificial Intelligence, Auckland, New Zealand, 1-5 December 2008, p. 337-347-
dc.identifier.isbn978-3-540-89377-6-
dc.identifier.urihttp://hdl.handle.net/10722/223760-
dc.description.abstractIn this paper, a new classification method (ADCC) for high dimensional data is proposed. In this method, a decision cluster classification model (DCC) consists of a set of disjoint decision clusters, each labeled with a dominant class that determines the class of new objects falling in the cluster. A cluster tree is first generated from a training data set by recursively calling a variable weighting <em>k</em> -means algorithm. Then, the DCC model is selected from the tree. Anderson-Darling test is used to determine the stopping condition of the tree growing. A series of experiments on both synthetic and real data sets have shown that the new classification method (ADCC) performed better in accuracy and scalability than the existing methods of <em>k</em> -<em>NN</em> , decision tree and SVM. It is particularly suitable for large, high dimensional data with many classes.-
dc.languageeng-
dc.publisherSpringer-Verlag Berlin.-
dc.relation.ispartofAI '08 Proceedings of the 21st Australasian Joint Conference on Artificial Intelligence: Advances in Artificial Intelligence-
dc.titleBuilding a Decision Cluster Classification Model for High Dimensional Data by a Variable Weighting k-Means Method-
dc.typeConference_Paper-
dc.identifier.emailHuang, JZ: jhuang@eti.hku.hk-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-540-89378-3_33-
dc.identifier.hkuros164906-
dc.identifier.spage337-
dc.identifier.epage347-
dc.publisher.placeGermany-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats