A new model of self-organizing neural networks and its application in data projection

@article{citeulike:2802802, abstract = {In this paper a new model of self-organizing neural networks is proposed. An algorithm called \“double self-organizing feature map\” (DSOM) algorithm is developed to train the novel model. By the DSOM algorithm the network will adaptively adjust its network structure during the learning phase so as to make neurons responding to similar stimulus have similar weight vectors and spatially move nearer to each other at the same time. The final network structure allows us to visualize high-dimensional data as a two dimensional scatter plot. The resulting representations allow a straightforward analysis of the inherent structure of clusters within the input data. One high-dimensional data set is used to test the effectiveness of the proposed neural networks}, author = {Su, Mu-Chun and Chang, Hsiao-Te }, booktitle = {Neural Networks, IEEE Transactions on}, citeulike-article-id = {2802802}, doi = {10.1109/72.896805}, journal = {Neural Networks, IEEE Transactions on}, keywords = {dsom, network, neural}, number = {1}, pages = {153--158}, posted-at = {2008-05-16 00:16:09}, priority = {2}, title = {A new model of self-organizing neural networks and its application in data projection}, url = {http://dx.doi.org/10.1109/72.896805}, volume = {12}, year = {2001} }

TY - JOUR ID - citeulike:2802802 L3 - citeulike-article-id:2802802 TI - A new model of self-organizing neural networks and its application in data projection T2 - Neural Networks, IEEE Transactions on JF - Neural Networks, IEEE Transactions on VL - 12 IS - 1 SP - 153 EP - 158 N2 - In this paper a new model of self-organizing neural networks is proposed. An algorithm called “double self-organizing feature map” (DSOM) algorithm is developed to train the novel model. By the DSOM algorithm the network will adaptively adjust its network structure during the learning phase so as to make neurons responding to similar stimulus have similar weight vectors and spatially move nearer to each other at the same time. The final network structure allows us to visualize high-dimensional data as a two dimensional scatter plot. The resulting representations allow a straightforward analysis of the inherent structure of clusters within the input data. One high-dimensional data set is used to test the effectiveness of the proposed neural networks KW - dsom KW - network KW - neural AU - Su, Mu-Chun AU - Chang, Hsiao-Te PY - 2001/// UR - http://dx.doi.org/10.1109/72.896805 ER -