For a college project, I need to create a dataset that includes multiple bugs, with each bug having a bug report, GitHub commit, and Stack Overflow post associated with it. I’m looking for methods to link these elements together, considering similarity scoring techniques like BM25. Any additional suggestions?

For a college project I need to find easy ways to create a dataset of bugs (Deep learning/ machine learning orientated) that includes bug reports from bugzilla, github and relevent posts from stackoverflow and other forums
I found that i can use similarity scoring methods like BM25 yet i am looking for more alternatives