GitEvolve: Predicting the Evolution of GitHub Repositories

Abstract

Software development is becoming increasingly open and collaborative with the advent of platforms such as GitHub. Given its crucial role, there is a need to better understand and model the dynamics of GitHub as a social platform. Previous work has mostly considered the dynamics of traditional social networking sites like Twitter and Facebook. We propose GitEvolve, a system to predict the evolution of GitHub repositories and the different ways by which users interact with them. To this end, we develop an end-to-end multi-task sequential deep neural network that given some seed events, simultaneously predicts which user-group is next going to interact with a given repository, what the type of the interaction is, and when it happens. To facilitate learning, we use graph based representation learning to encode relationship between repositories. We map users to groups by modelling common interests to better predict popularity and to generalize to unseen users during inference. We introduce an artificial event type to better model varying levels of activity of repositories in the dataset. The proposed multi-task architecture is generic and can be extended to model information diffusion in other social networks. In a series of experiments, we demonstrate the effectiveness of the proposed model, using multiple metrics and baselines. Qualitative analysis of the model's ability to predict popularity and forecast trends proves its applicability 1 . CCS CONCEPTS• Information systems → Social networks; • Computing methodologies → Neural networks; • Human-centered computing → Social networks.

Survivability Prediction of Open Source Software with Polynomial Regression

“…In contrast, our research models survivability as a continuum, enabling more detailed predictions about a project's health and long-term survival. Zhou et al predicts the future state of a repository by analyzing its evolutionary trends, employing a deep learning-based analysis of time-series data and predicting activities like star counts and commits, requiring substantial data collection [9]. Our method, focusing solely on bug-fixing activities, requires less data and is less complex.…”

“…Consequently, software maintenance costs may rise, prompting users to consider long-term support and maintenance when choosing OSS. Previous studies have focused on OSS development activities or version support [8][9][10]. However, unlike these methods, we conducted research focusing on the maintenance status and survivability of OSS based on the activities of contributors and users.…”

“…Therefore, it is necessary to evaluate the maintainability of OSS based on factors such as repository characteristics and current status and make informed choices. Previous studies have attempted to estimate OSS survivability by predicting the repository activity or estimating decreases in contributor numbers [8][9][10]. However, these studies focused on overall development activities, including project dependencies, which limited their ability to accurately assess the remaining risks related to bugs.…”

“…Zhou et al [9] offer significant insights into the survivability and sustainability of projects by predicting the development and evolution of repositories. The continuous evolution of a repository indicates the project's potential for sustainability, thereby facilitating the identification of its survivability.…”