What is it about?
If they want to be convincing, when data scientists explain their understanding of the data, they also have to explain how they came to that conclusion. Current data science tools don't make it easy for data scientists to track their process. We decided to track their process for them and label their steps to make their process easier to rationalize and communicate.
Featured Image
Why is it important?
Better communication can enable more data scientists to partake in deeply reflecting on analysis process and decisions of other data scientists. This reflection can lead to a better understanding of data by a whole community rather than one individual.
Perspectives
I think this work demonstrates the ability of ML tools to help us reflect on our own habits and processes in data science. I felt that much of what we uncovered could be applied towards helping us understand ourselves as humans.
Deepthi Raghunandan
University of Maryland at College Park
Read the Original
This page is a summary of: Code Code Evolution: Understanding How People Change Data Science Notebooks Over Time, April 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3544548.3580997.
You can read the full text:
Contributors
The following have contributed to this page
