Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jiang, MF | en_US |
dc.contributor.author | Tseng, SS | en_US |
dc.contributor.author | Su, CM | en_US |
dc.date.accessioned | 2014-12-08T15:43:53Z | - |
dc.date.available | 2014-12-08T15:43:53Z | - |
dc.date.issued | 2001-05-01 | en_US |
dc.identifier.issn | 0167-8655 | en_US |
dc.identifier.uri | http://dx.doi.org/10.1016/S0167-8655(00)00131-8 | en_US |
dc.identifier.uri | http://hdl.handle.net/11536/29677 | - |
dc.description.abstract | In this paper, a two-phase clustering algorithm for outliers detection is proposed. Tn;e first modify the traditional k-means algorithm in Phase 1 by using a heuristic "if one new input pattern is far enough away from all clusters centers, then assign it as a new cluster center". It results that the data points in the same cluster may be most likely all outliers or all non-outliers. And then we construct a minimum spanning tree (MST) in Phase 2 and remove the longest edge. The small clusters, the tree with less number of nodes, are selected and regarded as outlier. The experimental results show that our process works well. (C) 2001 Elsevier Science B.V. All rights reserved. | en_US |
dc.language.iso | en_US | en_US |
dc.subject | outliers | en_US |
dc.subject | k-means clustering | en_US |
dc.subject | two-phase clustering | en_US |
dc.subject | MST | en_US |
dc.title | Two-phase clustering process for outliers detection | en_US |
dc.type | Article | en_US |
dc.identifier.doi | 10.1016/S0167-8655(00)00131-8 | en_US |
dc.identifier.journal | PATTERN RECOGNITION LETTERS | en_US |
dc.citation.volume | 22 | en_US |
dc.citation.issue | 6-7 | en_US |
dc.citation.spage | 691 | en_US |
dc.citation.epage | 700 | en_US |
dc.contributor.department | 資訊工程學系 | zh_TW |
dc.contributor.department | Department of Computer Science | en_US |
dc.identifier.wosnumber | WOS:000168354200009 | - |
dc.citation.woscount | 63 | - |
Appears in Collections: | Articles |
Files in This Item:
If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.