Predicting Stars on Open-Source GitHub Projects

With rising popularity of platforms like GitHub, statistics like the number of stars can help comprehend the popularity of a project. In this paper, we evaluate a dataset of 2000 projects along with multiple GitHub features. The dataset informed using the GHTorrent project along with manual adjustments. Next, we build a smart model to predict the number of stars for a repository based on the features. Our work shows that it is indeed possible to predict the popularity/GitHub stars for a project based on the features. We use algorithms like Random Forest and a log-scaled dataset to achieve an R-square score of 0.88.