Student appeals analysis, from days to hours

After completing the Postgraduate Diploma in Data Science at Data Science Institute, Edward Chisoro streamlined student appeals analysis from days to hours using Python and NLP at the University of Johannesburg.

Edward is showing what happens when applied data science meets a real operational pain point. As Teaching & Learning Coordinator at the University of Johannesburg, in the School of Management (College of Business and Economics, Department of Finance and Investment Management), Edward had already taught himself Excel to compile student data, produce reports, and create visualisations. He also supports the liaison between students, faculty, and management and sits in committee meetings for student appeals and progression, where decisions must be both evidence-based and human.

For a long time, that process relied heavily on manual paperwork and basic compilation. The result: slow turnaround and inconsistent handling of complex cases, especially when appeals included lengthy written statements.

After completing the Postgraduate Diploma in Data Science with the Data Science Institute, Edward transformed the workflow.

This is what "data science in the real world" looks like: faster decisions, better consistency, and processes that become more humane because they're finally scalable.

Using Python, improved data management, and better-quality pipelines, he moved the appeals analysis from something that could take days or even weeks to a process that can be completed in a matter of hours with a huge uplift in efficiency and fairness.

He didn't stop at structured data like results, attendance, and progression trends. Edward also applied text mining and natural language processing to the unstructured side of appeals, analysing student narratives, surfacing themes, and even using word clouds to help committees quickly see patterns across large volumes of written submissions.

The impact has been recognised across the university and UJ is now going to fund Edward's Master's research, with growing interest from other schools within the College of Business and Economics in adopting the approach.

This is what "data science in the real world" looks like: faster decisions, better consistency, and processes that become more humane because they're finally scalable.