Automatic Learning Path Recommendation for Open Source Projects Using Deep Learning on Knowledge Graphs

Open source is an important way for developers to collaborate on software development. More and more developers begin contributing to open-source projects. When a developer begins to contribute to an existing open source project, the first thing to do is to read and understand the project code. However, most current open source projects only provide API documentation, not project design documents for new developers. Developers can only understand the code based on scattered comments in the code, which are difficult for new comers. Therefore, developers need to find a learning path, which helps them understand the project and finish their contribution tasks quickly. In order to help developers find the learning path easily and quickly, this paper puts forward a method to automatically recommend learning paths of open source projects. It uses multiple data sources in an open source community to extract knowledge data and build knowledge graphs for open source projects. After that, based on a deep-learning-based knowledge graph embedding model and a path recommendation algorithm, the method recommends proper learning paths for developers. We select three well-known open source projects, including Lua, Memcached and TensorFlow, according to language, scope and community activity, as cases to verify our method, and do comparative experiments between the learning paths found by real developers and recommended by the method. Experiment results show that our method saves developers a lot of time while ensuring the accuracy of the recommended learning path.