B-2-5 The calculation of significant degree of the data component in all kinds of distinction data by the Spherical Self Organizing-Maps (SSOM) method

Abstract

Using the iris data of the benchmark, the significant degree of each component was calculated by the Spherical Self-Organizing-Maps (SSOM) method and then the usual Plane SOM. First of all the algorithm of the significant degree was constructed and then, the method was applied to the iris data. The result was compared with the multiple regression analytical results. The results are fur better than the multiple regression analytical method. The method was then applied to other several benchmark data. The result could be quite reasonable. Also, it was applied to the fatigue questionnaire data.