For the large text categorization project at work, I developed tools that measure classification metrics such as precision and recall for our linguistic rules. These tools work across all sources and allow evaluation of changes our developers make both within a category and more broadly across the project.

As this project spans several hundred categories, I also developed a dashboard in Tableau that allows us to monitor the performance over time within each level of the classification hierarchy. It has been invaluable for internal meetings for executive business interests, clearly demonstrating our progress towards refining the rules to reach targets.