Enhancing Code Smell Classification with Code Refactoring-Based Data Augmentation Data Set

This repository contains the datasets used in the study “Enhancing Code Smell Classification with Code Refactoring-Based Data Augmentation.” Specifically, it includes: The original (standard) dataset, A rule-based augmented dataset generated through refactoring transformations, An LLM-based augmented dataset created using large language models. The primary purpose of this repository is to publicly release these datasets in order to support reproducibility, enable rigorous benchmarking, and facilitate future research in code smell classification and software quality assessment.本仓库包含用于《基于代码重构的数据增强提升代码异味（Code Smell）分类性能》这一研究的数据集。具体而言，本仓库包含以下数据集： 原始（标准）数据集； 基于规则的增强数据集，该数据集通过代码重构转换生成； 基于大语言模型（Large Language Model，LLM）的增强数据集，该数据集通过大语言模型生成。 本仓库的核心目的为公开发布上述数据集，以支持研究成果可复现性、开展严谨的基准测试，并推动代码异味分类与软件质量评估领域的后续研究工作。