计算机应用研究2024，Vol.41Issue(11)：3420-3425,6.DOI:10.19734/j.issn.1001-3695.2024.03.0085
基于知识回放的即时软件缺陷预测增量模型
Using knowledge replay for just-in-time software defect prediction incremental model
摘要
Abstract
Just-in-time software defect prediction technology enables just-in-time defect prediction at the granularity of code changes,which is crucial for improving software code quality and ensuring software reliability.Traditional static software de-fect prediction models suffer from'knowledge forgetting'when processing just-in-time software data streams,leading to poor model generalization performance.To address this,this paper proposed an incremental model method based on knowledge re-play for just-in-time software defect prediction.Firstly,it used the knowledge replay mechanism stores model parameters and random samples to facilitate the learning of old knowledge.Secondly,this paper used a distributed training framework to per-form incremental learning on just-in-time software data streams on local devices,achieving real-time model updates through re-structuring.Lastly,this paper employed the knowledge distillation technique to construct a global incremental prediction mo-del.Experiments show that this model performs better in terms of comprehensive performance compared to common modeling algorithms while ensuring training efficiency.关键词
即时软件缺陷预测/增量学习/知识回放/知识蒸馏Key words
just-in-time software defect prediction/incremental learning/knowledge replay/knowledge distillation分类
信息技术与安全科学引用本文复制引用
张文静,李勇,王越..基于知识回放的即时软件缺陷预测增量模型[J].计算机应用研究,2024,41(11):3420-3425,6.基金项目
新疆维吾尔自治区自然科学基金资助项目(2022D01A225) （2022D01A225）
国家自然科学基金资助项目(62241209) （62241209）
新疆师范大学研究生科研创新项目(XSY202301006) （XSY202301006）
