Visualizing and comparing multivariate scalar data over a geographic map.

Abstract

Recent technological advances and innovations have given us ways to easily and quickly extract large sets of data, but the increasing amounts of raw information only highlight the lack of good visualization or pattern recognition techniques to interpret the data. The objective of the research is to build techniques to effectively visualize multivariate scalar entities over a topographical map. Our goals are; a. rapid interpretation of the magnitude of a scalar entity at a particular spatial location; b. rapid comparison of the magnitudes of different scalar entities ; c. rapid comparison of the scalar entities across different regions of the map. Based on past research, I chose to investigate creating a texture of symmetrical units called texels. Each texel contains a fixed number of color-mapped hexagonal blocks representing each scalar entity. Users can dynamically choose the static variables to be visualized and the size of the texels. The research started as an experiment to visualize the United States Election results to represent the degree of variation in the results and the votes shared among the contestants. In addition to the election data my technique has also been applied to the United States census data, geographical and meteorological data to highlight interesting results.