CovParser is a personal project that I developed in school for computational linguistics. It comprises of a Securities and Exchange Commission (SEC) EDGAR webscraping tool and a covenant parsing tool for identifying sections within credit agreements.
The tools are written in Python and use webscraping, data analysis, regular expressions, natural language processing, and machine learning. The resulting document objects and models are reusable and extensible. The website itself is written from the ground up with Flask, has its own MySQL database and workers, and uses Python, MySQL, JavaScript, CSS, and HTML.
CovParser is part of my larger website project called Shotokan Scholar, a website for Shotokan karate. It, too, is a work in progress, but that means it's always growing!