Assignment 1 and 2: CFRP

The tool we’d build would be an interactive timeline with each show encapsulated inside a bubble color-coded by the genre of the show, the size of the bubble proportional to the number of tickets sold. Users can hover their cursor over the timeline to get exact ticket numbers, the name of the performance, and other details.

It would look something like this:

**Audience and Research Questions Addressed** This tool would be useful for historians studying what genres the french public were most interested in watching across history- this could reflect the mood of the public, for instance growing unrest may be reflected by a run of popular tragedies.

Implementation

**Data Extraction** To go from facsimile to data, we would use computer vision algorithms to process the images and extract text, then use a translation tool like Google Translate to convert the text to English. These automated methods are cheap, fast and scalable - nowadays many readily available libraries exist to detect writing in images and parse it to text (for example https://www.mathworks.com/help/vision/examples/automatically-detect-and-recognize-text-in-natural-images.html). The facsimiles contain quite simple language- mostly nouns and numbers- without any complex grammar, so it should be straightforward to translate. However, using machine vision may not be quite as accurate and reliable as typing up the pages by eye. There is such a large quantity of facsimiles that it would not be practical to do this, so computer vision is nevertheless the best method to use for this task.

**Data Storage and Presentation** We can extract the names of the performances, the number of tickets sold, and the annual income generated by the shows from JSON files. We can write a software program that directly take these pieces of info as input. This program would process the data and output an organized version of the data. Using this data, we organize the info onto a timeline in a html template and display the timeline on a webapp. Lastly, we can deploy the webapp into the Cloud.