Data Clustering Using Evidence Accumulation Export

@inproceedings{citeulike:3732716, abstract = {We explore the idea of evidence accumulation for combining the results of multiple clusterings. Initially, n d−dimensional data is decomposed into a large number of compact clusters; the K-means algorithm performs this decomposition, with several clusterings obtained by N random initializations of the K-means. Taking the cooccurrences of pairs of patterns in the same cluster as votes for their association, the data partitions are mapped into a co-association matrix of patterns. This n × n matrix represents a new similarity measure between patterns. The final clusters are obtained by applying a MST-based clustering algorithm on this matrix. Results on both synthetic and real data show the ability of the method to identify arbitrary shaped clusters in multidimensional data. 1.}, author = {Fred, Ana L. N.}, booktitle = {In Proc. of the 16th Int'l Conference on Pattern Recognition}, citeulike-article-id = {3732716}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.2.3429}, keywords = {clustering, data-mining}, pages = {276--280}, posted-at = {2008-12-01 13:52:45}, priority = {2}, title = {Data Clustering Using Evidence Accumulation}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.2.3429}, year = {2002} }

TY - CONF ID - citeulike:3732716 L3 - citeulike-article-id:3732716 N2 - We explore the idea of evidence accumulation for combining the results of multiple clusterings. Initially, n d−dimensional data is decomposed into a large number of compact clusters; the K-means algorithm performs this decomposition, with several clusterings obtained by N random initializations of the K-means. Taking the cooccurrences of pairs of patterns in the same cluster as votes for their association, the data partitions are mapped into a co-association matrix of patterns. This n × n matrix represents a new similarity measure between patterns. The final clusters are obtained by applying a MST-based clustering algorithm on this matrix. Results on both synthetic and real data show the ability of the method to identify arbitrary shaped clusters in multidimensional data. 1. EP - 280 TI - Data Clustering Using Evidence Accumulation T2 - In Proc. of the 16th Int’l Conference on Pattern Recognition SP - 276 KW - clustering KW - data-mining AU - Fred, Ana PY - 2002/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.2.3429 ER -