HIGHLY ACCURATE DATA CLASSIFICATION APPARATUS BASED ON TREE DIAGRAM CREATION UTILIZING SPHERICAL SELF-ORGANIZATION MAP AND PROGRAM FOR CREATING THE HIGHLY ACCURATE DATA CLASSIFICATION APPARATUS

摘要

PPROBLEM TO BE SOLVED: To display the whole spherical surface by respective colors included in respective data groups, by discriminating the data groups based on a tree diagram created along mutual positional relationship after end of learning of self-organization maps of reference vectors displayed on a spherical self-organization map and based on positional relationship on the spherical surface between data and respective nodes. PSOLUTION: A spherical self-organization map corresponding to a plurality of reference vectors is created, the code book vectors of the map are stored and considered, and nodes and input data are combined and stuck to the spherical surface, to generate tree diagram data by means of an already acquired patent. The order of the tree diagram data are stored, the order of the integrated data and the three-dimensional coordinates of respective nodes are checked, the proximity relationship of data labels on the tree diagram are checked, and respective nodes are discriminated and classified by colors. Furthermore, nodes to which labels belong may be immediately classified from the tree diagram created by combining node groups and data label groups, by considering the data label groups belonging to the tree diagram. PCOPYRIGHT: (C)2011,JPO&INPIT