Aging-related bugs and software complexity metrics



Change Log

When What
August 08th, 2015 Donated by Roberto Natella


Studies who have been using the data (in any form) are required to include the following reference:

  title={Predicting Aging-Related Bugs using Software Complexity Metrics},
  author={Cotroneo, Domenico and Natella, Roberto and Pietrantuono, Roberto},
  journal={Performance Evaluation},

About the data

This dataset contains information on aging-related bugs found in two open-source projects (the Linux kernel and the MySQL DBMS). This dataset has been used to investigate defect prediction approaches for aging-related bugs, by using software complexity metrics and machine learning techniques. New software complexity metrics were proposed in this study to support defect prediction (“aging-related” metrics).

The dataset contains an ARFF file for each subsystem of the open-source projects. Each row of the ARFF file contains:

  • The name of a file in the project;

  • “Program size” metrics for the file (columns from 2 to 50);

  • “McCabe’s cyclomatic complexity” metrics for the file (columns from 51 to 68);

  • “Halstead” metrics for the file (columns from 69 to 77);

  • “Aging-related” metrics for the file (columns from 78 to 83);

  • The number of aging-related bugs found in the file.