Disaster Response Classifier

Author: Heng Li

Summary:

This project is aimed at constructing a web-based pipeline to classify messages potentially related to disaster response. If related, the message is further classified into an individual category, such as 'Request', 'Food', 'Water', etc. This web-based application and underlying machine learning classifier can facilitate disaster response task by deciding whether a message from sources, such as social media and news, is related. It can help government and non-governmental organizations to quickly identify needs from areas and people struck by disasters.

Instructions:

1. Access web app deployed at Heroku at:
   https://response-classifier-webapp.herokuapp.com/

2. File structure of project repository:

   ├── data
      ├── DisasterResponse.db       # SQLite database containing clean data
      ├── disaster_categories.csv         # raw data containing disaster catergories
      ├── disaster_messages.csv        # raw data containing disaster messages
      └── process_data.py        # ELT pipeline
      
   ├── models
      ├── classifier.pkl         # trained classifier
      └── train_classifier.py          # machine learning pipeline to train and store classifier
      
   ├── app
      ├── run.py        # Flask file that runs app
      └── templates
          ├── go.html         # query and classfication result of web app
          └── master.html        # main page of web app
          
   ├── images
      ├── classifying_example.png         # screenshot of query classification result page
      └── index_page.png         # screenshot of main web page
   
   ├── Procfile         # Procfile to run app in Heroku
   ├── requirements.txt       # requirements on packages to be installed to run the web app
   └── utils.py         # utility functions

3. To run application locally:

   • Clone the git repository.

   • Run the following commands in the project's root directory to set up your 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 data/DisasterResponse.db

   • To run ML pipeline that trains classifier and saves python models/train_classifier.py data/DisasterResponse.db models/classifier.pkl

   • Run the following command in the app's directory to run your web app. python run.py

   • Go to http://0.0.0.0:3001/

4. Example screenshots of index page and classification:
   

   Index Page

   

   Classification Page

   

Acknowledgement:

This project was made possible because of the dataset provided by Figure Eight and training from Udacity Data Scientist Nanodegree program.