DevRank: Mining Influential Developers In Github

摘要

As the social coding is becoming increasingly popular, understanding theinfluence of developers can benefit various applications, such as advertisementfor new projects and innovations. However, most existing works have focusedonly on ranking influential nodes in non-weighted and homogeneous networks,which are not able to transfer proper importance scores to the real importantnode. To rank developers in Github, we define developer's influence on thecapacity of attracting attention which can be measured by the number offollowers obtained in the future. We further defined a new method, DevRank,which ranks the developers by influence propagation through heterogeneousnetwork constructed according to user behaviors, including "commit" and"follow". Our experiment compares the performance between DevRank and someother link analysis algorithms, the results have shown that DevRank can improvethe ranking accuracy.