Leveraging Race Prediction Algorithms to Enhance Team Composition in Big Data Science Teams

Abstract

As big data science projects scale in complexity, optimizing team composition has become vital for improving creativity, productivity, and project success. We explore the possibility of incorporating race prediction algorithms for enhancing racial diversity in team composition in big data science projects. This paper evaluates five race prediction algorithms—wru, ethnicolr, ethnicolr2, pyethnicity, and rethnicity—and then discuss their potential in supporting racially diverse team assembly in big data projects. Utilizing three datasets, we assess algorithm performance and applicability, emphasizing their role in building balanced teams that enhance agility, inclusivity, and bias mitigation. We present an actionable methodology for integrating demographic insights into team management. In addition, we propose ethical safeguards to ensure responsible race prediction use, recommending data privacy measures, aggregate-only data handling, and transparency in communication. We argue that when used within ethical constraints, race prediction can support robust team processes, reduce reliance on less diverse teams, and ultimately facilitate more creative and equitable big data project outcomes.