Disaster Response Pipeline Project (Udacity - Data Science Nanodegree)

Table of Contents

1. Description
2. Getting Started
   • Dependencies
   • Installing
   • Executing Program
3. Additional Material
4. Authors
5. License
6. Acknowledgement
7. Screenshots

Description

This project is part of the Data Science Nanodegree Program by Udacity. It involves building a Natural Language Processing (NLP) model to categorize messages from real-life disaster events in real-time. The dataset consists of pre-labelled tweets and messages.

The project is divided into the following key sections:

• Processing data: Building an ETL pipeline to extract, clean, and store the data in a SQLite database.
• Building a machine learning pipeline: Training a classifier to categorize text messages into various categories.
• Running a web app: Displaying model results in real-time.

Getting Started

Dependencies

• Python 3.5+
• NumPy, SciPy, Pandas, Scikit-Learn
• NLTK
• SQLAlchemy
• Pickle
• Flask, Plotly

Installing

Clone the git repository:

Executing Program

1. Run the ETL pipeline to clean data and store the processed data in the database:

   python data/process_data.py data/disaster_messages.csv data/disaster_categories.csv data/DisasterResponse.db
   

2. Run the ML pipeline to load data from the database, train the classifier, and save the classifier as a pickle file:

   python models/train_classifier.py data/DisasterResponse.db models/classifier.pkl
   

3. Run the web app:

   python run.py
   

   Access the web app at:

• Running on http://127.0.0.1:3001
• Running on http://192.168.29.170:3001

Authors

M S Mohan Kumar

License

This project is licensed under the MIT License.

Acknowledgement

• Udacity for providing an excellent Data Science Nanodegree Program.

Screenshots

• Sample Input
• Sample Output
• Main Page
• Process Data
• Train Classifier without Category Level Precision Recall