What refactoring detection tools can I use to meaningfully measure refactoring in python repositories? [closed]

Working on a research paper for a uni data mining repository course. The paper aims to explore technical debt in gen-ai code vs human-written code. Im focusing on: How do software developers refactor AI-assisted code in software repositories compared to human-written code? My dataset consists of python only repositories of small-medium size. I've used PyRef to get method level refactoring by commit but I'm struggling to see how this is meaningful to the research question and the paper. Using PyRef for each repo I've found Total Refactorings and Average Time-to-Refactor (sec), and all the refactorings done with the metrics. I wanted to try MLRefScanner but the authors repo seems like the tool isn't optimized for public usage yet. Any suggestions?

Mar 20, 2025 - 01:20

Working on a research paper for a uni data mining repository course. The paper aims to explore technical debt in gen-ai code vs human-written code.

Im focusing on: How do software developers refactor AI-assisted code in software repositories compared to human-written code?

My dataset consists of python only repositories of small-medium size. I've used PyRef to get method level refactoring by commit but I'm struggling to see how this is meaningful to the research question and the paper. Using PyRef for each repo I've found Total Refactorings and Average Time-to-Refactor (sec), and all the refactorings done with the metrics. I wanted to try MLRefScanner but the authors repo seems like the tool isn't optimized for public usage yet.

Any suggestions?