DACOS
收藏资源简介:
DACOS是由达尔豪斯大学哈利法克斯分校创建的手动标注代码异味数据集,包含10,267个针对5,192个代码片段的标注。该数据集专注于三种代码异味:多面抽象、复杂方法和长参数列表,通过两阶段过程创建,首先识别潜在主观性的代码片段,然后收集这些片段的标注。DACOS旨在帮助研究人员构建与上下文相关的机器学习模型,以检测代码异味,解决代码异味检测中的主观性问题。
DACOS is a manually annotated code smell dataset developed by Dalhousie University's Halifax campus, containing 10,267 annotations for 5,192 code snippets. This dataset focuses on three types of code smells: Multi-faceted Abstraction (Blob Class), Complex Method, and Long Parameter List. It was created through a two-stage process: first identifying code snippets with potential subjectivity, then collecting annotations for these snippets. DACOS aims to assist researchers in building context-aware machine learning models for code smell detection, addressing the subjectivity challenge inherent in code smell detection tasks.




