SOMを応用した軟組縁側貌分析に基づく顎変形症診断(<一般講演>自己組織化マップ(2)・他,医療情報システムにおけるソフトコンピューティグ)

抄録

In this paper, a new method of deciding the coefficient of each element of vector for obtaining reasonable classification results of vectors is described. A self-organizing map is useful for classification of vectors, however, it is not usable when the vectors including the difference of significance between elements. It is effective that the coefficients of the elements based on their significance is employed. In the proposed method, the coefficients of elements are decided by using genetic algorithm, in which the fitness is calculated by the classification results of the self-organizing map. The effectiveness and validity of the proposed method is verified by applying it to classification of the facial profiles.