This is a simple web app to read disaster response messages and judge what kind of messages they are.
- Using Logistic Regression to judge whether each message is categorized to each of 36 categories
- There are 36 models working parallel, and a message can be categorized in multiple categories
- As for the preprocessing, Tf-idf is applied
-
Run the following commands in
deploy
directory to set up the database and model.- To run ETL pipeline that cleans data and stores in database
python data/process_data.py data/disaster_messages.csv data/disaster_categories.csv sqlite:///data/database.db
- To run ML pipeline that trains classifier and saves
python models/train_classifier.py sqlite:///data/database.db models/finalized_model.pkl
- To run ETL pipeline that cleans data and stores in database
-
Run the following command in
deploy/app
directory to run the web app.python run.py
-
Go to http://localhost:3001/
-
Input a sentence on the input box and click the button below, then the analysis results will appear.