Predicting Fault-Prone Modules Using the Length of Identifiers

Author

Kawamoto, Kimiaki ; Mizuno, Osamu

Author_Institution

Grad. Sch. of Sci. & Technol., Kyoto Inst. of Technol., Kyoto, Japan

fYear

2012

fDate

26-27 Oct. 2012

Firstpage

30

Lastpage

34

Abstract

Identifiers such as variable names and function names in source code are essential information to understand code. The naming for identifiers affects on code understandability, thus, we expect that they affect on software quality. In this study, we examine the relationship between the length of identifiers and existence of software faults in a software module. The results of experiment using the random forest technique show that there is a positive relationship between the length of identifier and existence of software faults.

Keywords

random processes; software fault tolerance; software quality; source coding; code understandability; fault prone module prediction; identifier length; random forest technique; software fault; software module; software quality; source code; Accuracy; Computer bugs; Length measurement; Predictive models; Software algorithms; Software quality; Fault-prone module; Identifiers; Random forest; Repositories Mining;

fLanguage

English

Publisher

ieee

Conference_Titel

Empirical Software Engineering in Practice (IWESEP), 2012 Fourth International Workshop on

Conference_Location

Osaka

Print_ISBN

978-1-4673-4366-4

Type

conf

DOI

10.1109/IWESEP.2012.15

Filename

6363293