rehan-ahmar/message-classification

Multilabel message classification pipeline

Disaster Response Pipeline Project - Multi-label Classification of Messages

Table of Contents

1. Project Motivation
2. Installation
3. File Structure
4. Usage Instructions
5. Output
6. Acknowledgements

Project Motivation

This project was created as part of Udacity's Data Scientist for Enterprise nanodegree.

The main objectives of this project were the following:

• An ETL pipeline for loading data from csv files, transforming and cleaning the data, and finally storing it to an sqlite database.
• A Machine learning pipeline to predict the categories of different messages recorded during natural disasters.
• A web app that displays few visualisations of the data and classifies new input messages into multiple categories.

Installation

The code should run using any Python versions 3.*. I used python 3.6.

Libraries Used : numpy, pandas, sklearn, nltk, sqlalchemy, plotly, flask, bootstrap

File Structure

- app
| - template
| |- master.html  # main page of web app
| |- go.html  # classification result page of web app
| - run.py  # Flask file that runs app, contains code for visualization and prediction.

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

- models
|- train_classifier.py   # Machine Learning Pipeline
|- classifier.pkl  # saved model

- notebooks   # contains Jupyter notebooks used to create ETL and ML Pipelines.
- README.md

Usage Instructions

1. 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`

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

3. Go to http://localhost:3001

Output

In the web app, user can input a new message and get the predicted categories.

The web app also shows some visualizations for the training data in database.

Acknowledgements

Thanks to Figure Eight for the data, and Udacity for course material.