Note 2: This repo has been archived.

Note: A reference for dealing with JSON transcripts can be found in this code.

This project runs several types of text analysis on the corpus of messages from a GroupMe chat.

• Project Proposal; also in this repo
• Project Write-up

• Python 3.7 (except for the scripts in the utilities folder, which use 2.7).
• Make sure your transcript is up-to-date (see utilities/README.md)

Finally, set the path by typing the following into your command prompt:

export PROJECT_PATH="[path to this repo]"

Python 2.7

• README
• groupme-fetch.py: script for getting GroupMe transcripts
• newest-id.py: used by groupme-fetch.py
• simple-transcript.py: prints a human-readable version of a given transcript; same functionality can be found in clustering/Glaeser_assignment02.py
• stopwords.txt: standard list of stopwords (provided by Dr. Duncan Buell)

Python 3.7

Cluster users by the content of their messages.

• Write-up explaining original code
• Python code

Python 3.7

Determine the sentence complexity of each user's messages.

• Write-up explaining original code
• Glaeser_assignment03.ipynb and Glaeser_assignment03.py: full code

There are also files containing submodules of the full code:

• read_sentences.ipynb
• parse_sentences.ipynb
• results.ipynb

Python 3.7

Analyze the mood of the messages in the chat over time.

Python 3.7

Obtain topics that describe the content of the chat in a certain time frame.