Developer Project Recommendation Model Based on CNN-LSTM in GitHub

被引：0
作者：
Liao Z.-F. [1 ]
Yang H.-Y. [1 ,2 ]
Song T.-H. [2 ,3 ]
Yu S. [1 ]
Qi X.-F. [1 ]
机构：
[1] School of Computer Science and Engineering, Central South University, Changsha
[2] China Aeronautical Radio Electronics Research Institute, Shanghai
[3] School of Software, Central South University, Changsha
来源：
Tien Tzu Hsueh Pao/Acta Electronica Sinica | 2020年 / 48卷 / 11期
关键词：
GitHub; Project forecast; Project recommendation;
D O I：
10.3969/j.issn.0372-2112.2020.11.015
中图分类号：
学科分类号：
摘要：
As an open source project hosting platform, GitHub participates in the development of open source projects with multi-developers.As the core element of GitHub, developers ensure the activity of the whole system.However, many new projects can not find suitable collaborative developers in a short time and the development cycle gets delayed.To solve this problem, this paper proposes a CNN-LSTM developer project recommendation model based on Word2Vec, which trains developers to access the project sequence by Word2Vec, and vectorizes the project, calculates the project similarity with CNN-LSTM model, and recommends the appropriate project sequence for developers.Through the project prediction and similar project discovery experiments based on 62, 031 developers' project access data in GitHub in 2015, the experimental results show that the model has better recommendation effect and can help developers find similar projects of interest. © 2020, Chinese Institute of Electronics. All right reserved.
引用
收藏
页码：2202 / 2207
页数：5
相关论文
共 13 条
  • [1] Liu C, Yang D, Zhang X, Et al., Recommending GitHub projects for developer onboarding, IEEE Access, pp. 1-1, (2018)
  • [2] Thung F, Bissyande T F, Lo D, Et al., Network structure of social coding in GitHub, Proceedings of the 2013 European Conference on Software Maintenance and Reengineering, pp. 323-326, (2013)
  • [3] Liao Z, Jin H, Li Y, Et al., DevRank: Mining influential developers in Github, Proceedings of the 2017 IEEE Global Communications Conference, pp. 1-6, (2017)
  • [4] Rahman M M, Roy C K., An insight into the pull requests of GitHub, Proceedings of the 11th Working Conference on Mining Software Repositories, pp. 364-367, (2014)
  • [5] Rahman M M, Roy C K, Collins J A., CORRECT: Code reviewer recommendation in GitHub based on bross-project and technology experience, Proceedings of the 38th International Conference on Software Engineering Companion, pp. 222-231, (2016)
  • [6] Zhao G, Da C, Et al., Improving the pull requests review process using learning-to-rank algorithms, Empirical Software Engineering, 24, 4, pp. 2140-2170, (2019)
  • [7] Liao Zhi-fang, Liu Min, Et al., Markov chain-like model for prediction service based on improved hierarchical particle swarm optimization cluster algorithm, International Journal of Software Engineering & Knowledge Engineering, 26, pp. 653-674, (2016)
  • [8] Zhu Jing-hua, Et al., Multiple objects tracking algorithm by trajectories clustering in sensor network, Acta Electronica Sinica, 45, 11, pp. 2671-2676, (2017)
  • [9] Xu W, Sun X, Hu J, Et al., REPERSP: Recommending personalized software projects on GitHub, Proceedings of the 2017 IEEE International Conference on Software Maintenance and Evolution, pp. 648-652, (2017)
  • [10] Zhang X, Wang T, Yin G, Et al., Who will be interested in? A contributor recommendation approach for open source projects, Proceedings of the 29th International Conference on Software Engineering and Knowledge Engineering, pp. 363-369, (2017)
12