This project sets out to analyze disaster data from Figure Eight to build a model for an API that classifies disaster messages. The data set contains real messages that were sent during disaster events. The project produces a machine learning pipeline to categorize these events to help match the messages with an appropriate disaster relief agency.The project also includes a web app where an emergency worker can input a new message and get classification results in several categories. The web app also displays visualizations of the data.

The data comes from Figure Eight: https://www.figure-eight.com/

Two comma delimited text file (csv) are used in the project: messages and categories datasets. In the data preparation stage, the messages and categories datasets are merged using the common ID. The merged dataset is then cleaned through the following steps:

• Split the values in the categories column on the ; character so that each value becomes a separate column
• Use the first row of categories dataframe to create column names for the categories data
• Rename columns of categories with new column names
• Iterate through the category columns in df to keep only the last character of each string (the 1 or 0)
• Convert the string to a numeric value
• Drop the categories column from the df dataframe
• Concatenate df and categories data frames
• Drop the duplicates
• Store the cleaned dataset in a SQLite database

A machine learning pipeline is created in train_classifier.py, which performs the following tasks:

• 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

The final machine learning model chosen for production uses a Decision Tree model, which produces an accuracy rate of 94.6% on the test dataset. This model performed better compared to a random forest classifier, which produced an accuracy rate of 94.1% on the test dataset.

• 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 |- InsertDatabaseName.db # database to save clean data to

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

• README.md

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://0.0.0.0:3001/

Author Umrbek Allakulov