The CHum Word Visualization was our first experiment in D3. Our objective was to visualize token frequency over time primarily as a means to test the capabilities of D3. The visualization includes the full text of five hundred articles, minus stop words, selected from the Computers and the Humanities journal. We selected each article because it dealt with tools or the usage of tools. This is the same dataset used in the dendrogram visualization.
Each data point represents a relative frequency. That is the total number of times each token was counted in a year divided by the total number of counted tokens. All data was preprocessed in an R script; the smoothing provided by R's loess function.
Project Lead: Geoffrey Rockwell
Programming: Ryan Chartier