This is an experimental Python script that will convert JSON outputs from courtlistener.com into a directory of text files. This directory will then be suitable for ingesting into topic-modeling software such as the Stanford Natural Language Processing Group's Topic Modeling Tool.
install the lxml module
pip install lxml
To use, ensure that you have a directory of json files. Then, while in the same top-directory that contains the folder of JSON files, run jsontotext.py in the command line. The result will be a folder of opinion text files.
If you have any questions or comments please contact me!