Self Organizing Map and Sammon Mapping for Asymmetric Proximities Export

@incollection{citeulike:2239817, abstract = {Self Organizing Maps (SOM) and Sammon Mapping (SM) are two information visualization techniques widely used in the data mining community. These techniques assume that the similarity matrix for the data set under consideration is symmetric. However there are many interesting problems where asymmetric proximities arise, like text mining problems are. In this work we propose modified versions of SOM and SM to deal with data where the proximity matrix is asymmetric. The algorithms are tested using a real document database, and performance is reported using appropriate measures. As a result, the asymmetric algorithms proposed outperform their symmetric counterparts.}, author = {Martin-Merino, Manuel and Mu\~{n}oz, Alberto}, citeulike-article-id = {2239817}, citeulike-linkout-0 = {http://dx.doi.org/10.1007/3-540-44668-0_60}, doi = {10.1007/3-540-44668-0_60}, journal = {Artificial Neural Networks — ICANN 2001}, keywords = {asymetric, similarity, visualanalytics}, pages = {429--435}, posted-at = {2008-01-16 15:48:48}, priority = {2}, title = {Self Organizing Map and Sammon Mapping for Asymmetric Proximities}, url = {http://dx.doi.org/10.1007/3-540-44668-0_60}, year = {2001} }

TY - CHAP ID - citeulike:2239817 L3 - citeulike-article-id:2239817 N2 - Self Organizing Maps (SOM) and Sammon Mapping (SM) are two information visualization techniques widely used in the data mining community. These techniques assume that the similarity matrix for the data set under consideration is symmetric. However there are many interesting problems where asymmetric proximities arise, like text mining problems are. In this work we propose modified versions of SOM and SM to deal with data where the proximity matrix is asymmetric. The algorithms are tested using a real document database, and performance is reported using appropriate measures. As a result, the asymmetric algorithms proposed outperform their symmetric counterparts. JF - Artificial Neural Networks — ICANN 2001 EP - 435 TI - Self Organizing Map and Sammon Mapping for Asymmetric Proximities SP - 429 KW - asymetric KW - similarity KW - visualanalytics AU - Martin-Merino, Manuel AU - Muñoz, Alberto PY - 2001/// UR - http://dx.doi.org/10.1007/3-540-44668-0_60 ER -