%0 PDF %T An Analytical Approach for the Creative Design of New Visualizations. %A Griffin, Garth.; Li, Shaomeng.; Gramazio, Connor.; Chang, Remco. %D 2017-11-16T12:05:37.956-05:00 %8 2017-11-16 %I Tufts University. Tisch Library. %R http://localhost/files/73666g96n %X Educators have long understood the power of learning by example. Lee et al. made a compelling case for the efficacy of designing by example. However, designing by example is often approached in an ad-hoc fashion. Such opportunistic tactics may overlook promising avenues that would readily be discovered using a systematic method. Phylogenetic analysis has been successfully utilized in a number of domains for identifying previously unknown or undiscovered data points based on a corpus of example input data. In light of this success, we propose a method for using phylogenetic analysis of existing visualizations to discover novel visualizations in a systematic example-based creative design process. We algorithmically map existing visualizations into a phylogenetic tree based on their similarity to one another in terms of visual attributes. We use the phylogeny to find potential novel visualizations that are related to the existing examples, and we use the visual attributes of the existing examples to compute descriptions of these potential new designs. This represents a quantitative and systematic approach for example-based visualization design. In this paper, we briefly describe our approach, present a prototype system for visualization design using this method, and demonstrate the success of our technique through a sample usage scenario of the system. %[ 2018-10-09 %9 Text %~ Tufts Digital Library %W Institution