The Study on the Performance of Clustering Algorithms in the Determination of Combat Fatigues

전투복 표준호수를 결정하기 위한 클러스터링 알고리즘의 성능에 관한 연구

This paper evaluates the performance of clustering algorithms in the determination of combat fatigues. Among clustering algorithms, GLVQ in neural clustering methods. and ISODATA algorithm are evaluated in this paper. In order to evaluate, this study used the data sets which showed the body measurement of Korean military personnel. For the comparison of the three methods, the coverage rate and fitness for combat fatigues are evaluated and the convergence of final standard sizes are considered. As a result, the mean of the coverage rate and fitness for combat fatigues obtained with the GLVQ are higher than those obtained with -Means and ISODATA algorithms.