Classification of Medical Datasets Using Optimal Feature Selection Method with Multi-support Vector Machine

Abstract

AbstractAutomated Diagnosis in healthcare is becoming an interesting study in recent time among the data scientists to predict and diagnose the conditions in patients. In this manner, analysis of plays a major part in detection and classification of disease and accurately diagnose the medical condition in patients. Most of the data mining task is held up with poor classification accuracy due to the presence of redundant or irrelevant data items. In this research, the issue of poor classification accuracy is addressed and is solved by developing a framework that involves a series of stages. This includes pre-processing, feature extraction and classification of data items. The study uses Optimal Feature Selection Method (OFSM) as its feature selection tool and Multi-Support Vector Machine as its classification tool. The experimental validation is carried out to study the efficacy of the proposed method over various datasets and the outputs are evaluated in terms of accuracy, specificity, sensitivity and f-measure.KeywordsMulti-Support Vector MachineClassificationChronic Kidney DiseaseDEOFSM