The Fall quarter has been incredibly tight in terms of schedule but that has not deterred me from trying to apply the passion of data analysis on art history.
I started off with the plan of collecting thousands of samples throughout past decades and then using an API I found to do image analysis. After a lot of work, the API did not seem to provide me with the right tools so I switched to a version of the same software built to run through the terminal. In the process, I was able to refine my command line knowledge and produce a CSV output of all the samples I had collected. After performing some data wrangling on it through R, I have got the desired data in the form that I want. The moment when I got the correct data frame gave me happiness that one does not get in completing academic assignments.
Now I am working to find ways in which I can represent the data and make it informative through visuals. I thought that Tableau would be a good way to do that but after talking to my professor, we came to the conclusion that I should go with ggplot to convey my results. Hoping to complete this project soon and showcasing it to my former art history professors and the department as a tool that might be used as a reference for teaching art history. Also, planning on doing textual analysis of Shakespeare's plays after this project in an attempt to get to know more about his characters and themes.
Below is a snap of a part of the data frame I got from the analysis.
Update: Starting from the 1400s, I have collected 70 artworks and done analysis on them. Surprisingly, came up against a number of challenges after having got results for sample images with the biggest one being taking into account the accented characters that artworks' names consisted of and matching them with regex. The second image below is a snap of the preliminary analysis of the median of the blue color of RGB color space present in different artworks.