Applied visual analytics for exploring the National Health and Nutrition Examination Survey

Abstract

The National Health and Nutrition Examination Survey (NHANES) is a research program to assess the health and nutritional status of the population in the United States. In this work, we present a visual analytics system designed to help researchers explore patterns and form hypotheses within the NHANES dataset. The visualization component of the environment is an extension of traditional scatterplot matrices. Since the upper portion of the scatterplot matrix is a redundant encoding, we utilize this space, to show the projected N-dimensional clustering of points. The rows and columns of the matrix are automatically ordered using information about the cluster projection in each space as a means of showing the most meaningful dimensions. A comparison module has also been included that allows the user to compare groupings of people to the 2010 Dietary Guidelines for Americans. This tool enhances the analysis work by aiding discovery and hypothesis formation. © 2012 IEEE.