|
Journal of Korean Society for Quality Management 1994;22(2): 143-. |
집락분석과 판별분석의 활용성연구 |
채성산1, 황정연2 |
1대전대학교 이과대학 통계학과 2한국전자통신연구소 소프트웨어 종합검증실 |
Applicability of Cluster Analysis and Discriminant Analysis |
Seong-San Chae1, Jung-Yeon Hwang2 |
1Dept. of Statistics, Taejon University 2Electronics and Telecommunications Research Institute |
|
|
|
ABSTRACT |
Cluster analysis is a primitive technique in which no assumptions are made concerning the data structure. And the number of groups is known a priori discriminant analysis provides an information how well N individuals are classified into their own groups. In this study, clustering, which is any partition of a collection of data points, generated by the application of eight hierarchical clustering methods was re-classified by discriminant analysis. Then correct classification ratios were obtained for the application of discriminant analysis through each clustering method and the direct application of discriminant analysis. By comparing the correct classification ratios, the applicability of cluster analysis and discriminant analysis considered. |
|
|
|