A few years ago, Gary Albert and I were talking over beers about Charleston furniture and about features of furniture. Gary is a decorative arts historian, an Adjunct Curator at MESDA, the Museum of Early Southern Decorative Arts, and editor of their Journal of Early Southern Decorative Arts. I speculated that I might be able to do something quantitative with his feature dataset; at least generate hypotheses about relations between pieces and/or furniture makers.
Well, turns out I was able to do some useful cluster analysis. Very simple analysis on my part since the dataset was small and not that easy to quantify, but it turned out to be quite useful in hypothesis generation.
Here are the results, mostly Gary's work, but I did do the cluster analysis and help with an appendix. Very fun to try quantitative approaches with historical objects and craftsmanship.