My research concerns the intersection of natural language processing (NLP), human computer interaction, and systematic review. Systematic review is the process of identifying and synthesizing scientific publications to answer a specific scientific question. The general systematic review process is conducted by human reviewers and is time consuming, expensive, and prone to human error. Additionally, the rapid production of scientific publications makes the current human review practice unsustainable without the aid of NLP tools. My research concerns the development of interactive machine learning algorithms that quantify the variability and bias introduced by NLP in systematic review.