Survey on Classification and Feature Selection Approaches for Disease Diagnosis

Abstract

AbstractPatient case similarity implies that finding and extracting a patient case have similar features in the knowledge base. The knowledge base contains data obtained through demographics, progress notes, medications, past medical history, discharge summaries and lab values. Data pre-processing is the first step and an important step in the modelling process. The aim of this step is to increase the effectiveness of the classification process by using representative and consistent data set. Pre-processing includes data cleaning, data transformation and feature selection. Further, for predicting the new cases, new sample will be submitted to trained model. In the literature, various feature selection and classification approaches are available, but it is not clear which feature selection approach may have better classification performance. So, this study presents a survey on feature selection and classification approaches applied on seven benched-marked diseases data sets obtained from the UCI repository.KeywordsClassificationDisease diagnosisFeature selection