Semi-Automatic Content Analysis of Qualitative Data

Qualitative content analysis is commonly used by social scientists to understand the practices of the groups they study, but it is often infeasible to manually code a large text corpus within a reasonable time frame and budget. To address this problem, we are building a software tool to assist social scientists performing content analysis. We present our semi-automatic system that leverages natural language processing (NLP) and machine learning (ML) techniques for initial automatic coding, which human coders then review and correct. Through active learning, these human-verified annotations are subsequently used to train a higher performing model for machine annotation. We discuss design strategies adopted to optimize the system performance.