/static/assets/36.png

Evaluating a bot detection model on git commit messages

Mehdi Golzadeh
2021
0
Downloads
166
Views
0
Upvotes
Cite this Paper
0
Downloads
166
Views
0
Upvotes

Description

An automated tool to identify bots in Git repositories by analysing commit information. BoDeGiC accepts the name of a list of git repositories and computes its output in three steps. The first step consists of extracting all commit information from the specified git repositories using git log. This step results in a list of authors and their corresponding commits. The second step consists of computing the number of messages, empty messages, message patterns, and inequality between the number of messages within patterns. The third step simply applies the model we developed on these examples and outputs the prediction made by the model.
Terms of use

Comments