Abstract-The crowds' continuous participation and contribution are the key factors for the success of open source projects. However, among the massive competitors, it is difficult for a project to attract enough contributors by just passively waiting for enthusiasts to join in. Instead, it should actively seek gifted developers. Most of the current studies mainly focus on recommending experts inside a repository for some specific development tasks. In this paper, we propose a novel approach
A. IntroductionOpen source software (OSS) has become increasingly popular in software development. Quite different from the traditional software development, OSS is driven by massive crowds including developers, users, managers and so on. These stakeholders involve in OSS by interests, and most of them have their own full-time job and can only spend spare time on OSS. They can join in or withdraw from it at any time. Nevertheless, OSS has achieved great success at creating highquality software like Linux, MySQL, Spark and so on, and is viewed as "eating the software world" by "the Future of Open Source Survey" [1].The behavior of open source itself does not necessarily result in a project's success, but the continuous participation and active contribution from the crowds do play a crucial role. In GitHub, there are more than 48 million open source projects. However, over 2 million projects are not forked or watched by anyone after releasing, and 15.1% of them were not updated for more than a year and finally failed. Even DOI reference number: 10.18293/SEKE2017-067 for those projects which used to be successful will languish without continuous contribution of developers.Therefore, finding and attracting the right developers to participate is quite crucial for OSS. On the one hand, proper developers can provide necessary technical expertise that a project needs; On the other hand, suitable projects can inspire developers to participate in and to contribute their innovations continuously. However, there is a massive amount of competitive OSS, and developers are often limited by their time and energy to browse and choose from all the related projects. An automatic approach to bridge the gap between developers and projects and match them means much for both developers and projects.Several studies have explored the act of recommending developers to open source projects. Nguyen et al. [2] and Ma et al. [3] proposed approaches to fix the recommendation issues by analyzing the implementation history and expertise of developers. Thongtanunam et al. [4] and Yu et al. [5] focused on the automatic pull-request reviewer recommendation problem by analyzing the development history and social connections of individuals in the software community. Most of these studies mainly focus on specific software tasks and limit the recommended candidates to the core developers of the projects. By contrast, the current study is rooted in the granularity of open source repositories, and our output allows recommendations of external experts throughout the...