Abstract
When performing software evolution tasks, developers spend a significant amount of time looking for files to modify. By recommending files to modify, a code edit recommendation system reduces the developer’s navigation time when conducting software evolution tasks. In this paper, we propose a code edit recommendation method using a recurrent neural network (CERNN). CERNN forms contexts that maintain the sequence of developers’ interactions to recommend files to edit and stops recommendations when the first recommendation becomes incorrect for the given evolution task. We evaluated our method by comparing it with the state-of-the-art method MI-EA that was developed based on the association rule mining technique. The result shows that our proposed method improves the average recommendation accuracy by approximately 5% over MI-EA (0.64 vs. 0.59 F-score).
Highlights
Developers spend a significant amount of time looking for files to modify when performing software evolution tasks [1]
The Code Edit Recommendation Method Based on a Recurrent Neural Network we present our code edit recommendation method using a recurrent neural network (CERNN), which takes a context as an input and recommends multiple files to edit based on the context
The main difference between CERNN and MI-EA is that CERNN maintains sequential information related to developers’ operations and uses a deep learning technique, long short-term memory (LSTM)
Summary
Developers spend a significant amount of time looking for files to modify when performing software evolution tasks [1]. By recommending files to modify [2], a code edit recommendation system allows developers to reduce search time during software evolution tasks. To reduce the time spent by the developer on code navigation, Zimmermann et al [3] developed a recommendation system for mining association rules between changed files by collecting the revision history stored in the version management system. The recommendation system ROSE recommends files to edit with the context of one changed file and yields a 0.33 F-score [3]. Lee et al [1] developed a recommendation system MI-EA that works by mining association rules between viewed files and edited files from an interaction history, such as what was recorded by Mylyn [4] and stored in the Eclipse Bugzilla system. According to Lee et al [1], MI-EA yields a recommendation accuracy that is higher than that of ROSE because MI-EA uses a more elaborate context that includes viewed files
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
