Thesis (BS Applied Mathematics) -- University of the Philippines Mindanao, 2004

Several clustering algorithms have been suggested to do site analysis, but fewer solutions have been implemented to guide the design of clustering-based experiments and asses the quality of their outcomes. Cluster analysis aims to identify groups of similar objects and, therefore helps to discover distribution of patterns and interesting correlation in large data sets. A cluster validity framework provides insights into the problem of predicting what sites go together and determining the common and unique attributes that characterize the different cluster formed. These methods were applied to study the phenotypic diversity of feral cats in Mindanao. Evaluation approaches includes hierarchical clustering which yields the nested grouping of patterns and similarly levels at which grouping changes and the partitional (k-means) clustering which obtained a single partition of the data instead of a significantly supported site analysis application for the study of phenotypic diversity