Mehdi Golzadeh

Mehdi Golzadeh, Alexandre Decan, Tom Mens

Detecting the presence of bots in distributed software development activity is very important in order to prevent bias in large-scale socio-technical empirical analyses. In previous work, we proposed a classification model to detect bots in GitHub repositories based on the pull request and issue comments of GitHub accounts. The current study generalises the approach to git contributors based on their commit messages. We train and evaluate the classification model on a large dataset of 6,922 git contributors. The original model based on pull request and issue comments obtained a precision of 0.77 on this dataset. Retraining the classification model on git commit messages increased the precision to 0.80. As a proof-of-concept, we implemented this model in BoDeGiC, an open source command-line tool to detect bots in git repositories.

Mehdi Golzadeh, Alexandre Decan, Damien Legay et al.

Bots are frequently used in Github repositories to automate repetitive activities that are part of the distributed software development process. They communicate with human actors through comments. While detecting their presence is important for many reasons, no large and representative ground-truth dataset is available, nor are classification models to detect and validate bots on the basis of such a dataset. This paper proposes a ground-truth dataset, based on a manual analysis with high interrater agreement, of pull request and issue comments in 5,000 distinct Github accounts of which 527 have been identified as bots. Using this dataset we propose an automated classification model to detect bots, taking as main features the number of empty and non-empty comments of each account, the number of comment patterns, and the inequality between comments within comment patterns. We obtained a very high weighted average precision, recall and F1-score of 0.98 on a test set containing 40% of the data. We integrated the classification model into an open source command-line tool to allow practitioners to detect which accounts in a given Github repository actually correspond to bots.

Mehdi Golzadeh, Alexandre Decan, Eleni Constantinou et al.

Development bots are used on Github to automate repetitive activities. Such bots communicate with human actors via issue comments and pull request comments. Identifying such bot comments allows preventing bias in socio-technical studies related to software development. To automate their identification, we propose a classification model based on natural language processing. Starting from a balanced ground-truth dataset of 19,282 PR and issue comments, we encode the comments as vectors using a combination of the bag of words and TF-IDF techniques. We train a range of binary classifiers to predict the type of comment (human or bot) based on this vector representation. A multinomial Naive Bayes classifier provides the best results. Its performance on a test set containing 50% of the data achieves an average precision, recall, and F1 score of 0.88. Although the model shows a promising result on the pull request and issue comments, further work is required to generalize the model on other types of activities, like commit messages and code reviews.