The Game of Reviewing Security Clearances is the new and fun way to see the mistakes of others. Completely free to play, come on over and take a stab...just not at a human. You wont get a clearance that way.
First, I downloaded and parsed the data from http://ogc.osd.mil/doha/industrial/
Using the Python html parser, Beautiful Soup, I organized the security clearance data into a progrmattically-accessbile format. https://www.crummy.com/software/BeautifulSoup/bs4/doc/
Each adjudication has a digest which briefly describes the decision-making process for each adjudication. I looked for keywords in each digest which would clarify if an application for clearance was accepted or denied. Using the python Natural Language Toolkit, I parsed each adjudication for specific keywords, and stored the outputs in an sqlite3 database http://www.nltk.org/
Inputted the results from previous web scraping into the django backend sqlite database. Required previously created django models in order to correctly organize the data for usage in django.
Edited the admin panel to allow administrator to quickly edit data on the backend database. Can now edit adjudication details, decisions, and user data from the django admin GUI
Creating user interface with bootstrap http://django-bootstrap3.readthedocs.io/en/latest/quickstart.html
Implemented login system, form processing, and interactions between outout html templates and backend sqlite database
Manually removed obvious markers from each adjudication, reviewed each result and removed false answers from earlier language processing, removed a lot of debt-related ones because they got pretty boring after awhile
Packaged and deployed to AWS elastic beanstalk instance
SSHed into related AWS EC2 instance, installed Apache and mod_wsgi (httpd mod_wsgi)