The main aim of this study is to focus on the classification of homogeneous region and regional frequency analysis for the annual maximum daily rainfall in southern Taiwan. The principal component analysis has been used, in the beginning, to investigate the main components and its dispersion of annual maximum daily rainfall. The main components are then replaced the original components for achieving the purpose of data reduction and data interpretation. After examining the first several main components, the components which correlate in classification of probability distribution are chosen. These components have been employed in cluster analysis for identifying the rainfall areas with the same probability distribution. As a result, the A and B regions own the character of PT3 distribution, while the C and D regions have the character of GEV distribution.