jetBlue Sentiment Analysis

This project combines data scraped from twitter.com and airlinequality.com to determine user sentiment using machine learning

Comment

One of the driving factors that lead to this project idea was our desire to explore the intersection of machine learning and data visualization. Using machine learning we were able to rank data we parsed on a scale of -1 to 1 where 1 was very good sentiment and -1 was very bad sentiment. Building upon this, we created an interactive visualization using D3.js to better understand the relationship between when the data was posted onto the internet and the sentiment it received from the ml algorithm. This allowed us to see trends in the data that would otherwise have not been visible.

To run our project follow these steps: Prerequisites Python3 and Pip3.7 (or pip3)

Installing First, create a virtual environment

python3 -m virtualenv env After creating the virtual environment, start the environment using the following command.

source env/bin/activate Now, install all the dependencies as specified in requirements.txt

make init Be sure to set the environment variable for flask

export FLASK_APP=app.py Running the server To start the server, run the following command: flask run

Share this project:

Updates