Revisiting Method-Level Change Prediction: Comparative Evaluation at Different Granularities
To improve the efficiency of software maintenance, change prediction techniques have been proposed to predict modules that change frequently.While existing techniques primarily focus on class-level prediction, method-level prediction allows for more direct identification of change locations.Although method-level change prediction techniques have also been proposed, developers cannot decide when to use which one due to the lack of comparisons with class-level predictions.In this paper, we evaluated the performance of method-level change prediction in comparison with class-level prediction from three perspectives: direct comparison, method-level comparison , and maintenance effort-aware comparison.The results from 15 open source projects show that, although method-level prediction has lower performance than class-level prediction in usual evaluation, method-level prediction outperformed class-level prediction when both were evaluated at method-level, leading to a median difference of 0.26 in accuracy.Furthermore, effort-aware evaluation shows that method-level prediction had significantly better performance when maintenance effort is little.
Fri 7 MarDisplayed time zone: Eastern Time (US & Canada) change
11:00 - 12:30 | Change Management & Program ComprehensionReproducibility Studies and Negative Results (RENE) Track / Research Papers / Early Research Achievement (ERA) Track at L-1710 Chair(s): Masud Rahman Dalhousie University | ||
11:00 15mTalk | AdvFusion: Adapter-based Knowledge Transfer for Code Summarization on Code Language Models Research Papers Iman Saberi University of British Columbia Okanagan, Amirreza Esmaeili University of British Columbia, Fatemeh Hendijani Fard University of British Columbia, Chen Fuxiang University of Leicester | ||
11:15 15mTalk | EarlyPR: Early Prediction of Potential Pull-Requests from Forks Research Papers | ||
11:30 15mTalk | The Hidden Challenges of Merging: A Tool-Based Exploration Research Papers Luciana Gomes UFCG, Melina Mongiovi Federal University of Campina Grande, Brazil, Sabrina Souto UEPB, Everton L. G. Alves Federal University of Campina Grande | ||
11:45 7mTalk | On the Performance of Large Language Models for Code Change Intent Classification Early Research Achievement (ERA) Track Issam Oukay Department of Software and IT Engineering, ETS Montreal, University of Quebec, Montreal, Canada, Moataz Chouchen Department of Electrical and Computer Engineering, Concordia University, Montreal, Canada, Ali Ouni ETS Montreal, University of Quebec, Fatemeh Hendijani Fard University of British Columbia | ||
11:52 15mTalk | Revisiting Method-Level Change Prediction: Comparative Evaluation at Different Granularities Reproducibility Studies and Negative Results (RENE) Track Hiroto Sugimori School of Computing, Institute of Science Tokyo, Shinpei Hayashi Institute of Science Tokyo DOI Pre-print | ||