Disaster Response Classifier

Table of contents

1. Introduction
2. How to run
   1. Requirements
   2. File descriptions
   3. Instructions
3. Results
   1. Model Accuracy
   2. Classification
4. Acknowledgement

Introduction

Objective:

We would like to categorize real messages that were sent during disaster events so that we can send the messages to an appropriate disaster relief agency.

origin data:

Overview:

We will read the dataset provided by Appen, clean the data, and then store it in a SQLite database. We will then create a machine learning pipeline that uses NLTK, as well as scikit-learn's Pipeline and GridSearchCV to output a final model to predict classifications for 36 categories (multi-output classifications). Finally, there is a web app included where an emergency worker can input a new message and get classification results in several categories.

How to run

1. Requirements

Python 3
Libraries: numpy, pandas, sys, sqalchemy, re, NLTK, pickle, Sklearn, plotly and flask libraries

2. File descriptions

- 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          # Read, clean, and store data
|- InsertDatabaseName.db    # database to save clean data to

- models
|- train_classifier.py      # machine learning pipeline
|- classifier.pkl  　　　　　# saved model

- image                     # images of the data and results

- README.md

3. Instructions

1. Data cleaning
   
   Run the following command in the root directory of the project.

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

2. Model training
   
   Run the following command in the root directory of the project.

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

3. Starting the web app
   
   Go to the directory of the app and run the following command.

  'python run.py'

4. Go to http://0.0.0.0:3001/

Results

1. Model Accuracy

presicion: 0.78
recall: 0.38
f1-score: 0.48

2. Classification

Acknowledgement

Thanks to Appen for the data set and Udacity for the template and training.