An Online Developer Profiling Tool Based on Analysis of GitLab Repositories
An Online Developer Profiling Tool Based on Analysis of GitLab Repositories
复制标题
基于 GitLab 存储库分析的在线开发人员分析工具
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Hailong Sun
中科院分区：
文献类型：
--
作者：
Jing Wang;Xiangxin Meng;Huimin Wang;Hailong Sun
随着软件在当今经济社会发展中发挥着越来越重要的作用，开发者也越来越多。因此，科学地评估开发者的专业水平已成为互联网公司和开发者的迫切需求。但目前似乎还没有令人满意的方法来满足这一需求。在本文中，我们提出了一种通过分析开发人员源代码来分析开发人员的解决方案。我们基于GitLab代码库对开发人员的代码数量、代码质量、技能、贡献、个性化提交时间以及参与的项目进行分析。我们从代码数量、代码质量、贡献度和参与项目得分四个角度综合评估开发人员的专业知识。与现有方法相比，我们的评估指标更加全面。我们设计并实现了一个在线工具，可以提供开发人员搜索和分析。我们的工具已被东软和万达集团用来描述其软件开发人员的专业知识和绩效。
There are more and more developers as software is playing increasingly important roles in today’s economic and social development. As a result, evaluating developers’ expertise scientifically has become an urgent need for both Internet companies and developers. However, it seems that there is no satisfactory method to meet this demand currently. In this paper, we propose a solution to profile developers by analyzing their source code. We conduct the analysis of developers in terms of code quantity, code quality, skills, contribution, personalized commit time, and projects they participated in based on the GitLab code repositories. And we comprehensively evaluate developers’ expertise from four perspectives, which are code quantity, code quality, contribution and score of projects they participated in. Compared with existing methods, our evaluation indicators are more comprehensive. We design and implement an online tool that can provide developer searching and profiling. Our tool has been used in Neusoft and Wonders Group to characterize the expertise and performance of their software developers.