Recommended Practices for Python Pedagogy in Graduate Data Science Courses

This Research to Practice Full Paper discusses experiences and recommendations highlighting best practices for teaching Python¹ in data science at higher education institutions. Python has become the language of choice for teaching data science at both graduate and undergraduate programs. From data preprocessing, pattern extraction, predictive modeling, to visualization, the language has important libraries and tools to achieve this seamlessly. In this paper, the pedagogical approaches and learning objectives achieved by a cohort of graduate students specific to using Python at all stages of the KDD (Knowledge Discovery in Databases) life cycle are highlighted. The need to do this is motivated by describing an undergraduate data mining course taught at a STEM program in Computer Science. Teaching and research experiences to build projects are also elaborated across these groups. The paper then proposes recommended practices for teaching Python at the graduate level in similar data science programs.¹Python Software Foundation, [Online]. Available: https://www.python.org/