Affiliation:
1. Concordia University, Montreal, QC, Canada
2. Polytechnique Montréal, Montreal, QC, Canada
Abstract
Code embeddings have seen increasing applications in software engineering (SE) research and practice recently. Despite the advances in embedding techniques applied in SE research, one of the main challenges is their generalizability. A recent study finds that code embeddings may not be readily leveraged for the downstream tasks that the embeddings are not particularly trained for. Therefore, in this article, we propose
GraphCodeVec
, which represents the source code as graphs and leverages the Graph Convolutional Networks to learn more generalizable code embeddings in a task-agnostic manner. The edges in the graph representation are automatically constructed from the paths in the abstract syntax trees, and the nodes from the tokens in the source code. To evaluate the effectiveness of
GraphCodeVec
, we consider three downstream benchmark tasks (i.e., code comment generation, code authorship identification, and code clones detection) that are used in a prior benchmarking of code embeddings and add three new downstream tasks (i.e., source code classification, logging statements prediction, and software defect prediction), resulting in a total of six downstream tasks that are considered in our evaluation. For each downstream task, we apply the embeddings learned by
GraphCodeVec
and the embeddings learned from four baseline approaches and compare their respective performance. We find that
GraphCodeVec
outperforms all the baselines in five out of the six downstream tasks, and its performance is relatively stable across different tasks and datasets. In addition, we perform ablation experiments to understand the impacts of the training context (i.e., the graph context extracted from the abstract syntax trees) and the training model (i.e., the Graph Convolutional Networks) on the effectiveness of the generated embeddings. The results show that both the graph context and the Graph Convolutional Networks can benefit
GraphCodeVec
in producing high-quality embeddings for the downstream tasks, while the improvement by Graph Convolutional Networks is more robust across different downstream tasks and datasets. Our findings suggest that future research and practice may consider using graph-based deep learning methods to capture the structural information of the source code for SE tasks.
Publisher
Association for Computing Machinery (ACM)
Reference96 articles.
1. Large-Scale and Language-Oblivious Code Authorship Identification
2. Suggesting accurate method and class names
3. Miltiadis Allamanis, Marc Brockschmidt, and Mahmoud Khademi. 2018. Learning to represent programs with graphs. In Proceedings of the 6th International Conference on Learning Representations. OpenReview.net. Retrieved from https://openreview.net/forum?id=BJOFETxR-.
4. Miltiadis Allamanis, Hao Peng, and Charles A. Sutton. 2016. A convolutional attention network for extreme summarization of source code. In Proceedings of the 33rd International Conference on Machine Learning series. JMLR.org, 2091–2100. Retrieved from http://proceedings.mlr.press/v48/allamanis16.html.
5. A general path-based representation for predicting program properties
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. LoGenText-Plus
: Improving Neural Machine Translation Based Logging Texts Generation with Syntactic Templates;ACM Transactions on Software Engineering and Methodology;2023-12-22
2. Towards Utilizing Natural Language Processing Techniques to Assist in Software Engineering Tasks;2023 IEEE/ACM 45th International Conference on Software Engineering: Companion Proceedings (ICSE-Companion);2023-05
3. On the Temporal Relations between Logging and Code;2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE);2023-05