Georgetown Capstone Project: Cyberbullying

A repository for the Georgetown School of Continuing Studies Data Science Certificate capstone. Project is to identify instances of cyberbullying.

Project Structure

Project structure contains the following modules:

• analysis: Python code + Jupyter notebooks for analysis
• cron: Python code to setup various cronjobs e.g. daily Reddit scraper
• data: Example data sets for local storage (temporary; should be used for exploratory purposes; upload to AWS instead)
• examples: Example scripts for working with AWS RDS database + S3, Pandas, writing class wrapper for psycopg2
• papers: Papers collected that investigates the cyberbullying problem space

Daily Crons

This project was setup using the free tier of AWS. The EC2 instance is an Amazon instance, has Python 3.4 installed, and requires a Python 3.4 virtualenv for the daily cron to run as is. The cronjobs are ran as the default ec2-user.

Example for how to setup cron jobs:

1. ssh -i path/to/key.pem ec2-user@public-ec2-domain
2. Logic to activate Python 3.4 virtualenv
3. python cron/daily_crons.py
4. You can view cronjobs via crontab -e

Getting Started

To get started you'll need to:

• Setup a Python 3.4 virtualenv (by convention named venv as that's what's ignored in .gitignore) using the requirements.txt file, else use conda
• Note: If using AWS EC2 Amazon image, you may need to confirm the proper pip is used to install depedendencies. You don't want system pip but the virtualenv pip /venv/local/bin/pip3.4.
• Create a config.yml using your credentials, use example-config.yml as a template
• Setup AWS credentials via aws configure

Questions?

Contact Lorena Mesa via email (me@lorenamesa.com) or http://twitter.com/loooorenanicole.

Thanks folks!