Medical Document Classification from OHSUMED Dataset

Abstract

Investigation of biological databases is not straight forward and generically needs identification of Medical Document Classification (MDC) based on hierarchy and relationship between different strata (ontogeny). Thus, MDC remains as a challenging effort. Popularly, earlier text classification has applied flat classifier. However, our research aims to show the text classification in which we opt to assess the hierarchical organization of classes or categories. In order to fulfill the aim of our research, we are considering the human disease hierarchical structure of human disease ontology with the help of simple relation from biomedical text abstracts and the ontology learning. We conducted experiments to evaluate the effects of different representations by measuring the change in classification performance with MEDLINE documents from the OHSUMED dataset. This research suggest a hierarchical classification method employing the hierarchical concept structure for classifying biomedical text abstracts by using Hidden Markov Model method (HMM). Present study demonstrates how a large number of biomedical articles are divided into quite a few subgroups in a hierarchy describing ontogeny.