Λεπτομέρειες βιβλιογραφικής εγγραφής
| Τίτλος: |
DevRec: enhancing developer recommendation using machine learning. |
| Συγγραφείς: |
Al-Safoury, Laila, ElKorany, Abeer, Makady, Soha |
| Πηγή: |
Journal of King Saud University - Computer & Information Sciences; Jul2026, Vol. 38 Issue 5, p1-21, 21p |
| Θεματικοί όροι: |
Machine learning, Recommender systems, Open source software, Statistics, Computer programmers, Resource allocation, Feature extraction, Prediction models |
| Περίληψη: |
In Open-Source Software (OSS) projects, effectively managing tasks such as committing code, merging pull requests, and closing issues require accurate role-based recommendations for developers. As the number of issues and contributors increases, assigning the right developers to specific roles becomes more complex. This paper introduces DevRec, a novel framework designed to enhance multi-role Developer Recommendation for OSS projects. DevRec employs Semantic Analysis (SA) to interpret issue descriptions and historical developer contributions in OSS projects, thus improving the mapping between issues and suitable developers. Two complementary approaches are explored: Statistical-based Semantic Analysis (Statistical-SA) for ranking developer recommendations, and Machine Learning–Semantic Analysis (ML-SA) for predictive developer assignment. ML models are trained on previous developer behaviour and contributions across multiple projects to enhance recommendation accuracy. The framework is evaluated on a benchmark dataset comprising three OSS projects of varying sizes—Framework (small, 325 issues), Travis (medium, 5,457 issues), and Elasticsearch (large, 10,423 issues). Results show that Statistical-SA improves accuracy by 13.85%, 48.67%, and 1.66% respectively, over baseline methods. At the same time, ML-SA consistently outperforms Statistical-SA across all roles, achieving an average 2–5% improvement in Commit, Merge, and Close tasks. These findings highlight the effectiveness of integrating semantic and machine learning-based models for multi-role developer recommendations. [ABSTRACT FROM AUTHOR] |
|
Copyright of Journal of King Saud University - Computer & Information Sciences is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| Βάση Δεδομένων: |
Complementary Index |