Computer Science
8 years ago

New technique predicts future software quality, revolutionizing refactoring strategies.

On the use of time series and search based software engineering for refactoring recommendation
Hanzhang Wang, Marouane Kessentini, William I. Grosky, Haythem Meddeb

Paper Summary

Related papers

Theme background image
Computer Science
3 years ago

Developers Repay Technical Debt Through Refactoring, Improving Code Quality

Empirical Studies
Log Analysis
Information Flow Security
1 view
Theme background image
Computer Science
3 years ago

Refactoring code may lead to more faults, study finds.

Empirical Studies
Software Reliability
Log Analysis
Theme background image
Computer Science
3 years ago

Eliminating Code Smells Boosts Software Quality and Saves Maintenance Costs

Empirical Studies
Software Reliability
Software Testing
2 views

Refactoring is a way to improve software without changing how it works. This study suggests using a mix of time series and search-based methods to recommend the best ways to refactor software. By predicting how code quality will change in future releases, this approach helps manage technical debt more effectively. The researchers tested this method on real and open-source projects, showing it can lead to better refactoring decisions than other techniques.