Visualizing large data by the SOM and GTM methods - what are we obtaining?
نویسنده
چکیده
Visualizing large data by the SOM and GTM methods – what are we obtaining? Anna Bartkowiak University of Wrocław, Institute of Computer Science, Przesmyckiego 20, Wrocław 51–151, PL Abstract We consider the visualization of multivariate data yielded by the SOM and the GTM methods. Special emphasis is put on the position of outliers and extreme points. When evaluating four data sets, it is found that: The GTM method yields decidedly smaller quantization errors and much higher topological errors as the SOM does. Generally, the topology of both representations looks similar.
منابع مشابه
Generative topographic mapping applied to clustering and visualization of motor unit action potentials.
The identification and visualization of clusters formed by motor unit action potentials (MUAPs) is an essential step in investigations seeking to explain the control of the neuromuscular system. This work introduces the generative topographic mapping (GTM), a novel machine learning tool, for clustering of MUAPs, and also it extends the GTM technique to provide a way of visualizing MUAPs. The pe...
متن کاملSelf-organization and missing values in SOM and GTM
In this paper, we study fundamental properties of the Self-Organizing Map (SOM) and the Generative Topographic Mapping (GTM), ramifications of the initialization of the algorithms and properties of the algorithms in the presence of missing data. We show that the commonly used principal component analysis (PCA) initialization of the GTM does not guarantee good learning results with high-dimensio...
متن کاملVisualizing the Organization of the SOM and GTM
The Lattice of a Self-Organizing Map can be considered as an ”elastic network” that is fitted into the data space. The network is non-linear but locally the non-linear surface can be approximated by a linear hyperplane. This property could be utilized to visualize the direction of the network at any location of the map, by fitting a surface to the model vectors of the neighbors of a node and vi...
متن کاملExperimental Analysis of GTM
Not linear methods for statistical data analysis have become more and more popular thanks to the rapid development of computers. The fields in which they are applied to are as various as the methods them self. Generative topographic mapping (GTM) has been developed by [Bishop et al. 1997] as a principal alternative to the self-organizing map (SOM) algorithm [Kohonen 1982] in which a set of unla...
متن کاملNovel Visualisation Methods for Protein Data
Visualization of high-dimensional data has always been a challenging task. Here we discuss and propose variants of non-linear data projection methods (Generative Topographic Mapping (GTM) and GTM with simultaneous feature saliency (GTM-FS)) that are adapted to be effective on very highdimensional data. The adaptations use log space values at certain steps of the Expectation Maximization (EM) al...
متن کامل