Change-proneness datasets
- Citation Author(s):
-
Hadeel Alsolai (Princess Nourah Bint Abdulrahman University, Riyadh, Saudi Arabia)
- Submitted by:
- Hadeel Alsolai
- Last updated:
- DOI:
- 10.21227/m4kx-j760
123 views
- Categories:
- Keywords:
Abstract
I performed pre-processing methods on refactoring datasets proposed in (Empirical evaluation of software maintainability based on a manually validated refactoring dataset) by PéterHegedűs et al. The new version of these datasets support Change-proneness study.
Instructions:
The independent variables are numeric and include 125 metrics, which can be grouped into ten categories as follows: cohesion, complexity, coupling, documentation, inheritance, size, code duplication, warning, rules and refactoring. the dependent variable is Boolean ( True or False) and can be used to predict Change prone.
The datasets can be used to apply different machine learning models using different evaluation measures for classification problem.
