Be the Data: Embodied Interaction in Bayesian Visual Analytics
Be the Data promotes creative interactive data explora- tion through the novel combination of the physical and virtual worlds. The fundamental assertion is that data exploration, and the necessary understanding of the complex analytical methods behind data exploration, is stymied by its traditional strict limitation to the virtual world. Thus, we combined the physical and virtual to enable students to “be the data”. This means that a system was designed in which students can enter a physical space and embody virtual data points. Students can then physically move about the space to explore relationships among their data points. Their movement in the space is tracked so as to apply these physical interactions to the virtual mathematical models. In turn, students take an egocentric perspective on the data, interacting with each other in collaborative groups to explore interactions among data points.
Eighty 7th grade girls from southwestern Virginia learned about data analytics and visualization at the Women in Computing workshop today at Virginia Tech. In particular, they learned how to explore high-dimensional data by manipulating 2D projections of the data. They did this by "being the data" in the VT Cube. In groups of 20, the girls explored a data set containing 30 variables and about 20 different animals. Each girl embodied a single data point, one of the animals. Her position and movement in the Cube was tracked with respect to all the other girls. Together, the girls explored alternative clusterings of the data points by re-organizing themselves in the space. With the Andromeda software developed at VT, they then visualized the weighting of the data variables that produce their particular clustering. Through this mechanism the girls posed several research questions and hypotheses about the data and were able to discover new answers. With this interactive and collaborative approach, the girls also learned about the underlying mathematical concepts of high-dimensional data analysis.
The "Be the Data" project is a interdisciplinary collaborative research effort, exploring embodied analytics and education, led by Leanna House (Statistics), Chris North (Computer Science), Scotland Leman (Statistics), Jane Robertson (Statistics), and Liesl Baum (ICAT), and is supported by a seed grant from the Institute for Creativity, Arts, and Technology (ICAT). The Andromeda software was developed by graduate students Jessica Zeitz Self, Xin Chen, and Xinran Hu under an NSF supported project in the Discovery Analytics Center (DAC) at Virginia Tech.