Determining the factors that contribute to the healthcare disparity in United States is a substantial problem that healthcare professionals have confronted for decades. In this study, our objective is to build precise and accurate classification models to predict the factors, which attribute to the disparity in healthcare coverage in the United States. The study utilizes twenty-three variables and 67,636 records from the 2007 Behavioral Risk Factor Surveillance System (BRFSS). In our comparative analysis, three statistical feature extraction methods, Chi-Square, Gain Ratio, and Info Gain, were used to extract a set of relevant features, which were then subjected to the classification models, AdaBoost, Random Forest, Radial Basis Function (RBF), Logistic Regression, and Naïve Bayes, to analyze healthcare coverage. The most important factors that were discovered in the model are presented in this paper. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Sethi, P., & Jain, M. (2010). A Comparative feature selection approach for the prediction of healthcare coverage. Communications in Computer and Information Science, 54, 392–403. https://doi.org/10.1007/978-3-642-12035-0_41
Mendeley helps you to discover research relevant for your work.