"Class Name Recommendation based on Graph Embedding of Program Elements"

• Here is the repository for APSEC'19 paper
• The source code preprocessing the dataset is at RelationExtractor/src/main/java
• The source code of the proposed approach is at core
• All experiments in the paper are available in exp.ipynb

Requirements

• openjdk == 11.0.1
• python == 3.6.8
• pipenv version 2018.11.26
• maven == 3.6.2
• CUDA == 8.0.61

Procedure

1. clone this repository
2. cd APSEC2019Class
3. mkdir data | mkdir data/input data/model data/output data
4. download the data and put it in data/raw_data
5. javac RelationExtractor/src/main/java/FilePathOrganizer.java | java RelationExtractor/src/main/java/FilePathOrganizer
6. cd RelationExtractor | mvn clean install
7. cd .. | java -cp RelationExtractor/target/RelationExtractor-1.0-SNAPSHOT-jar-with-dependencies.jar Main
8. pipenv install
9. pipenv run jupyter notebook

BibTeX

@inproceedings{kurimoto2019class,
  title={Class Name Recommendation Based on Graph Embedding of Program Elements},
  author={Kurimoto, Shintaro and Hayase, Yasuhiro and Yonai, Hiroshi and Ito, Hiroyoshi and Kitagawa, Hiroyuki},
  booktitle={2019 26th Asia-Pacific Software Engineering Conference (APSEC)},
  pages={498--505},
  year={2019},
  organization={IEEE}
}