Philograph is an application built to assist literary scholars in their research. The application offers usability for analyzing connections in and between texts based on themes and clusters of keywords.

Makes use of the supervised machine-learning technique Support Vector Machines and the unsupervised clustering algorithm k-means.

Tailored for use in analyzing a corpus of sermons written by 18th-century Puritan minister, Jonathan Edwards. See sermons for the dataset used in this project.

The directory research-data contains results generated from this dataset.

Disclaimer: Philograph works best on macOS and has not been tested on other operating systems.

• Unix-like operating system (macOS or Linux)
• Basic understanding of command-line
• python should be installed
• pip3 should be installed. Philograph has many dependencies and it is likely you will have to install several packages, e.g. numpy or sklearn.

1. At the command prompt, git clone this repository (or download it as a zip archive).

    $ git clone https://github.com/jerrybonnell/philograph.git
   

2. Change directory to philograph and launch the application.

    $ cd philograph
    $ python philograph.py 
   

   You should see a help screen that displays suggested options. If not, there are likely missing packages. Use pip3 to install them.

• -p <corpus folder> --predict <corpus folder>

  launches predicting function, SVM

• -c <corpus folder> --cluster <corpus folder>

  launches clustering function, k-means

• -o <filename> --output <filename>

  writes results to a file with name <filename>

• -v --verbose

  runs in verbose mode. -vv runs with more verbosity and -vvv runs with highest verbosity.

• -h --help

  shows the help screen!

Note that the parameters used by these models can be customized by editing settings.json.

Developed as part of the CSC410/CSC411 project course at the University of Miami.

For further information regarding this project, please contact Jerry Bonnell .