A stability based method for discovering structure in clustered data. Export

@article{Ben2002, abstract = {We present a method for visually and quantitatively assessing the presence of structure in clustered data. The method exploits measurements of the stability of clustering solutions obtained by perturbing the data set. Stability is characterized by the distribution of pairwise similarities between clusterings obtained from sub samples of the data. High pairwise similarities indicate a stable clustering pattern. The method can be used with any clustering algorithm; it provides a means of rationally defining an optimum number of clusters, and can also detect the lack of structure in data. We show results on artificial and microarray data using a hierarchical clustering algorithm.}, address = {BioWulf Technologies LLC, 2030 Addison st. Suite 102, Berkeley, CA 94704, USA.}, author = {Ben-Hur, A. and Elisseeff, A. and Guyon, I.}, citeulike-article-id = {3063990}, citeulike-linkout-0 = {http://view.ncbi.nlm.nih.gov/pubmed/11928511}, citeulike-linkout-1 = {http://www.hubmed.org/display.cgi?uids=11928511}, issn = {1793-5091}, journal = {Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing}, keywords = {clustering, model\_select}, pages = {6--17}, posted-at = {2008-07-30 19:50:24}, priority = {2}, title = {A stability based method for discovering structure in clustered data.}, url = {http://view.ncbi.nlm.nih.gov/pubmed/11928511}, year = {2002} }

TY - JOUR ID - Ben2002 L3 - citeulike-article-id:3063990 N2 - We present a method for visually and quantitatively assessing the presence of structure in clustered data. The method exploits measurements of the stability of clustering solutions obtained by perturbing the data set. Stability is characterized by the distribution of pairwise similarities between clusterings obtained from sub samples of the data. High pairwise similarities indicate a stable clustering pattern. The method can be used with any clustering algorithm; it provides a means of rationally defining an optimum number of clusters, and can also detect the lack of structure in data. We show results on artificial and microarray data using a hierarchical clustering algorithm. JF - Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing SN - 1793-5091 AD - BioWulf Technologies LLC, 2030 Addison st. Suite 102, Berkeley, CA 94704, USA. EP - 17 TI - A stability based method for discovering structure in clustered data. SP - 6 KW - clustering KW - model_select AU - Ben-Hur, A AU - Elisseeff, A AU - Guyon, I PY - 2002/// UR - http://view.ncbi.nlm.nih.gov/pubmed/11928511 ER -