brwoodside / metis-neurips-twitter-proj-4 Goto Github PK

In this project, you'll learn how to extract data from Twitter for a specific search query and time frame, and then use the Twitter API to retrieve full text tweets. You'll then clean and parse this data, in order to perform dimensionality reduction and clustering to model topics about the corpus.

This exercise reveals that a large portion of conversation about the conference involves "hot takes" or reactions to ongoing events, praise for good workshops or speakers, or event announcements, like poster sessions.

Future work may involve gathering content based on search queries that involve "paper," "workshop," or "talk."

You can run these notebooks in order:

1. 1-Data-Acquistion.ipynb
2. 2-Data-Cleaning-NLP.ipynb
3. 3-Topic-Modeling.ipynb
4. 4-Visualization.ipynb
5. 5-Clustering.ipynb
6. 6-Embeddings.ipynb