Hierarchical Clustering Support Vector Machines for Classifying Type-2 Diabetes Patients

Using a large national health database, we propose an enhanced SVM-based model called Hierarchical Clustering Support Vector Machine (HCSVM)that utilizes multiple levels of clusters to classify patients diagnosed with type-2diabetes. Multiple HCSVMs are trained for clusters at different levels of the hierarchy.Some clusters at certain levels of the hierarchy capture more separablesample spaces than the others. As a result, HCSVMs at different levels may developdifferent classification capabilities. Since the locations of the superiorSVMs are data dependent, the HCSVM model in this study takes advantage ofan adaptive strategy to select the most suitable HCSVM for classifying the testingsamples. This model solves the large data set problem inherent with the traditionalsingle SVM model because the entire data set is partitioned into smallerand more homogenous clusters. Other approaches also use clustering and multipleSVM to solve the problem of large datasets. These approaches typical employedonly one level of clusters. However, a single level of clusters may notprovide an optimal partition of the sample space for SVM trainings. On the contrary,HCSVMs utilize multiple partitions available in a multilevel tree to capturea more separable sample space for SVM trainings. Compared with the traditionalsingle SVM model and one-level multiple SVMs model, the HCSVM Modelmarkedly improves the accuracy for classifying testing samples.