Static source code analysis is an increasingly important activity to manage software project quality, and is often found as a part of the development process. A widely adopted way of checking code quality is through the detection of violations to specific sets of rules addressing good programming practices. SonarQube is a platform able to detect these violations, called Issues. In this paper we described an empirical study performend on two industrial projects, where we used Issues extracted on different versions of the projects to predict changes in code through a set of machine learning models. We achieved good detection performances, especially when predicting changes in the next version. This result paves the way for future investigations of the interest in an industrial setting towards the prioritization of Issues management according to their impact on change-proneness.

Tollin, I., Arcelli Fontana, F., Zanoni, M., Roveda, R. (2017). Change prediction through coding rules violations. In ACM Proceedings of the 21th International Conference on Evaluation and Assessment in Software Engineering (EASE) (pp.61-64). Karlskrona : Association for Computing Machinery [10.1145/3084226.3084282].

Change prediction through coding rules violations

Arcelli Fontana F;Zanoni, M;Roveda, R
2017

Abstract

Static source code analysis is an increasingly important activity to manage software project quality, and is often found as a part of the development process. A widely adopted way of checking code quality is through the detection of violations to specific sets of rules addressing good programming practices. SonarQube is a platform able to detect these violations, called Issues. In this paper we described an empirical study performend on two industrial projects, where we used Issues extracted on different versions of the projects to predict changes in code through a set of machine learning models. We achieved good detection performances, especially when predicting changes in the next version. This result paves the way for future investigations of the interest in an industrial setting towards the prioritization of Issues management according to their impact on change-proneness.
slide + paper
software quality, change prediction, issues, machine learning.
English
Evaluation and Assessment in Software Engineering Conference (EASE) 15-16 June
2017
ACM Proceedings of the 21th International Conference on Evaluation and Assessment in Software Engineering (EASE)
9781450348041
2017
128635
61
64
open
Tollin, I., Arcelli Fontana, F., Zanoni, M., Roveda, R. (2017). Change prediction through coding rules violations. In ACM Proceedings of the 21th International Conference on Evaluation and Assessment in Software Engineering (EASE) (pp.61-64). Karlskrona : Association for Computing Machinery [10.1145/3084226.3084282].
File in questo prodotto:
File Dimensione Formato  
EASE_2017_paper_129.pdf

accesso aperto

Descrizione: ease 2017 proceeding
Dimensione 446.34 kB
Formato Adobe PDF
446.34 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/155474
Citazioni
  • Scopus 19
  • ???jsp.display-item.citation.isi??? ND
Social impact