-
Notifications
You must be signed in to change notification settings - Fork 2
/
README
15 lines (7 loc) · 615 Bytes
/
README
1
2
3
4
5
6
7
8
9
10
11
12
13
Predicting Popular Government, an analys is of states
By Justin Hines
This is a final project for Data Driven Modeling taught by Jake Hofman.
This project aims to predict the popularity of governemntal articles using a multi-class bernoulli naive bayes classifier.
Run make to download the dataset, filter, and apply the classifier. Results will be stored in log/. All tsv data sets will be in tsv organized by dstate. Final filtered tsvs will be in tsv/{STATE}/tsvs/
Run make bayes to run the bayes classifier on a dataset, or python bayes.py in the src/ directory.
The final report is included in docs/