● Re-engineered tagging tool designed to preprocess data for machine learning so as to better identify different entities of an educational research document
● Added functionality so that the tagging tool could process authors, section headers, and
footnotes to create training data for the machine learning algorithms
● Performed algorithmic data analysis of training data to identify unique properties of each
entity to improve accuracy of machine learning models
● Redesigned layout of SQL database used for machine learning workflow to improve
efficiency by storing previous output
● Implemented interface with Stanford Parser which sent and received JSON messages to add Natural Language Processing information to corpus data
● Created and designed a web based document management system for meeting minutes using ASP.NET Core, including the supporting SQL tables
● Created and designed keyword extraction algorithm for meeting minutes documents