基于决策树的代码异味优先级评估  

Decision tree based code smell prioritization evaluation

作　　者：郭迪 吴海涛[1] GUO Di;WU Haitao(College of Information,Mechanical and Electrical Engineering,Shanghai Normal University,Shanghai 201418,China)

机构地区：[1]上海师范大学信息与机电工程学院,上海201418

出　　处：《上海师范大学学报（自然科学版）》2022年第2期210-216,共7页Journal of Shanghai Normal University(Natural Sciences)

基　　金：国家自然科学基金(61672355)。

摘　　要：基于以开发人员驱动的代码异味优先级排序方法,结合优化决策树算法建立模型,对代码异味的重构优先级进行面向开发人员的排名,并在实证研究中评估了该模型,以模型可解释性方法对特征的重要性进行评估,给出了相关影响较高的特征.结果表明,该模型的F1值为89%,分别较基线值和最新研究成果高出25%和5%.Based on the prioritization method of developer-driven code smell,combining with the optimization decision tree algorithm,a model was built to rank the refactoring priority of code smell for developers,which was evaluated in the empirical study.The importance of features was evaluated by the interpretive methods and features with high correlation impact were provided.The experimental results showed that the F1 value of the model was 89%,which was 25%and 5%better than the benchmark value and the latest research result,respectively.

关 键 词：代码异味 决策树 特征选择 软件可维护性 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]