Mining Github for contributors at the Tensorflow repo & analysis of followship relations.
The development collaboration in open source project refers to the contribution of many developers to creating and enhancing a certain software. By being a development collaboration platform, and a social network that introduces the notion of followship, Github allows us to rise problematics around these two notions (followship and collaboration). In other words, given the contributors in a repository, we want to see if those contributors are connected, and how connected they are. All this can be done through social network analysis metrics and algorithms. We would also like to provide a graphical view of the output network and some insights from it.
So, the tools and packages used for this project are:
1- Data extraction: The Github API for python PyGithub (source below).
2- Network manipulation and graph analysis: NetworkX (Source below).
3- Network Visualization: Cytocscape (Source bellow).
PyGithub: https://pypi.org/project/PyGithub/
NetworkX: https://networkx.github.io/
Cytoscape: https://cytoscape.org/what_is_cytoscape.html