File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: A feature weighting approach to building classification models by interactive clustering

TitleA feature weighting approach to building classification models by interactive clustering
Authors
KeywordsFeature weight
Data mining
Clustering
Classification
DCC
Issue Date2004
PublisherSpringer.
Citation
First International Conference on Modeling Decisions for Artificial Intelligence (MDAI 2004), Barcelona, Spain, 2-4 August 2004. In Modeling Decisions for Artificial Intelligence: First International Conference, MDAI 2004, Barcelona, Spain, August 2-4, 2004: Proceedings, 2004, p. 284-294 How to Cite?
Abstract© Springer-Verlag Berlin Heidelberg 2004. In using a classified data set to test clustering algorithms, the data points in a class are considered as one cluster (or more than one) in space. In this paper we adopt this principle to build classification models through interactively clustering a training data set to construct a tree of clusters. The leaf clusters of the tree are selected as decision clusters to classify new data based on a distance function. We consider the feature weights in calculating the distances between a new object and the center of a decision cluster. The new algorithm, W-k-means, is used to automatically calculate the feature weights from the training data. The Fastmap technique is used to handle outliers in selecting decision clusters. This step increases the stability of the classifier. Experimental results on public domain data sets have shown that the models built using this clustering approach outperformed some popular classification algorithms.
Persistent Identifierhttp://hdl.handle.net/10722/276656
ISBN
ISSN
2005 Impact Factor: 0.302
2020 SCImago Journal Rankings: 0.249
Series/Report no.Lecture Notes in Computer Science ; 3131

 

DC FieldValueLanguage
dc.contributor.authorJing, Liping-
dc.contributor.authorHuang, Joshua-
dc.contributor.authorNg, Michael K.-
dc.contributor.authorRong, Hongqiang-
dc.date.accessioned2019-09-18T08:34:16Z-
dc.date.available2019-09-18T08:34:16Z-
dc.date.issued2004-
dc.identifier.citationFirst International Conference on Modeling Decisions for Artificial Intelligence (MDAI 2004), Barcelona, Spain, 2-4 August 2004. In Modeling Decisions for Artificial Intelligence: First International Conference, MDAI 2004, Barcelona, Spain, August 2-4, 2004: Proceedings, 2004, p. 284-294-
dc.identifier.isbn9783540225553-
dc.identifier.issn0302-9743-
dc.identifier.urihttp://hdl.handle.net/10722/276656-
dc.description.abstract© Springer-Verlag Berlin Heidelberg 2004. In using a classified data set to test clustering algorithms, the data points in a class are considered as one cluster (or more than one) in space. In this paper we adopt this principle to build classification models through interactively clustering a training data set to construct a tree of clusters. The leaf clusters of the tree are selected as decision clusters to classify new data based on a distance function. We consider the feature weights in calculating the distances between a new object and the center of a decision cluster. The new algorithm, W-k-means, is used to automatically calculate the feature weights from the training data. The Fastmap technique is used to handle outliers in selecting decision clusters. This step increases the stability of the classifier. Experimental results on public domain data sets have shown that the models built using this clustering approach outperformed some popular classification algorithms.-
dc.languageeng-
dc.publisherSpringer.-
dc.relation.ispartofModeling Decisions for Artificial Intelligence: First International Conference, MDAI 2004, Barcelona, Spain, August 2-4, 2004: Proceedings-
dc.relation.ispartofseriesLecture Notes in Computer Science ; 3131-
dc.subjectFeature weight-
dc.subjectData mining-
dc.subjectClustering-
dc.subjectClassification-
dc.subjectDCC-
dc.titleA feature weighting approach to building classification models by interactive clustering-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-540-27774-3_27-
dc.identifier.scopuseid_2-s2.0-9444284442-
dc.identifier.spage284-
dc.identifier.epage294-
dc.publisher.placeBerlin-
dc.identifier.issnl0302-9743-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats