Disaster Response Pipeline

---

1. Project Overview
2. Installation
3. Project Components
4. Results
5. Files
6. Acknowledgements

1. Project Overview

The aim of this project is to build a model for classifies disaster messages. The dataset collected by Figure Eight. The disaster response dataset contains 30,000 messages, It has been encoded with 36 different categories related to disaster response and has been stripped of messages with sensitive information in their entirety.

2. Installation

• Python versions 3.*.
• Python Libraries:
  • Pandas.
  • Scikit-learn.
  • numpy.
  • nltk.
  • sqlalchemy.

3. Project Components

There are three main folder in this project :

1. ETL Pipeline: data/process_data.py, contains data cleaning pipeline that:

   • Loads the messages and categories datasets
   • Merges the two datasets
   • Cleans the data
   • Stores it in a SQLite database

2. ML Pipeline: models/train_classifier.py contains machine learning pipeline that:

   • Loads data from the SQLite database
   • Splits the dataset into training and test sets
   • Builds a text processing and machine learning pipeline
   • Trains and tunes a model using GridSearchCV
   • Outputs results on the test set
   • Exports the final model as a pickle file

3. Flask Web App: contains web app to classifier messages in real time using trained model.

4. Results

Here are a few screenshots of the web app.

5. Files

- app
| - template
| |- master.html  # main page of web app
| |- go.html  # classification result page of web app
|- run.py  # Flask file that runs app

- data
|- disaster_categories.csv  # data to process 
|- disaster_messages.csv  # data to process
|- process_data.py
|- DisasterResponse.db   # database to save clean data to

- models
|- train_classifier.py
|- classifier.pkl  # saved model 

- README.md

6. Acknowledgements

I wish to thank Figure Eight for dataset. Also, thanks for Udacity for advice.