Wave Motion : A New Metaphor for 2 D Information Visualization The Exploration of a Metaphor

The use of conventional two or three-dimensional visualizations of quantitative information have an inherent limitation of only being able to present the interactions of 2 to 4 variables. In exploration of a new visual metaphor, we have drawn on the natural phenomena of wave motion. Specifically, the interference of waves was used as the underlying metaphor. In visualizing a set of data points, we sought to elicit the trends or patterns in the data set. In this project, the patterns are revealed in the form of interference between waves generated using the information in the data set. Just as stones dropped into water generate rings of ripples spreading outward to interact with each other, the data points are placed in a plane and each one generates a wave of a particular wave length, frequency, phase and amplitude. The resulting interference pattern can be used to detect the trends in a data set over six variables. The mapping of the data set variables to the two coordinates and four wave parameters is not automatic. Pending a better understanding of how the trends are manifested in the interference patterns, the assignment of variables to wave parameters has been done on an empirical basis. Artificial data sets were created using various distributions, such as normal and uniform, and various functions. The objective of using the artificial data sets was to isolate the possible patterns being generated. Software was developed to display the interference patterns for the data sets. The natural phenomenon of wave amplitude decay over distance was not implemented as this would bias the resulting pattern against the variable used for the amplitude, by a function of distance which is a combination of the variables used for the X and Y coordinates. Using the software, several experiments were performed: (1) to identify similarities between plots, which were created using data from the same artificial data set. (2) to identify deviation between plots which were created using two different artificial data sets. (3) to elicit more understanding of wave parameters by alternating the variables in the same artificial data set through several combinations of wave parameter assignments and comparing plots. The results obtained provide no distinctive patterns or features, which could be used reliably. Further research is needed to explore the metaphor's potential for pattern identification and data presentation.