Unsupervised learning plays an important role in knowledge exploration and discovery. Two basic examples of unsupervised learning are clustering and dimensionality reduction. In this paper, we introduce an improved model for clustering based on a hierarchical analysis method. In our model, there are three main steps. In the first step, we use a structural clustering model to find qualitative patterns from a given dataset. Then, the second step applies a quantitative-based clustering algorithm to find quantitative patterns from the dataset. The third and the last step generates hybrid patterns by combining the patterns obtained from the first two steps based on a certain criterion so that deeply hidden relationships can be extracted from the dataset. In this paper, we also discuss the results of our experiments with the proposed model and algorithms on longitudinal medical records.
Copyright 2006 IEEE. Reprinted from Proceedings of the 6th International conference on intelligent systems design and applications (ISDA 2006). This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of Macquarie University’s products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to firstname.lastname@example.org. By choosing to view this document, you agree to all provisions of the copyright laws protecting it.