Quality Assessment Based on Attribute Series of Software Evolution

Created by W.Langdon from gp-bibliography.bib Revision:1.4420

  author =       "Jacek Ratzinger and Harald Gall and Martin Pinzger",
  title =        "Quality Assessment Based on Attribute Series of
                 Software Evolution",
  booktitle =    "14th Working Conference on Reverse Engineering, WCRE
  year =         "2007",
  pages =        "80--89",
  address =      "Vancouver",
  month =        oct,
  publisher =    "IEEE",
  keywords =     "genetic algorithms, genetic programming, SBSE",
  isbn13 =       "978-0-7695-3034-5",
  annote =       "The Pennsylvania State University CiteSeerX Archives",
  bibsource =    "OAI-PMH server at citeseerx.ist.psu.edu",
  language =     "en",
  oai =          "oai:CiteSeerX.psu:",
  rights =       "Metadata may be used without restrictions as long as
                 the oai identifier remains attached to it.",
  URL =          "http://citeseerx.ist.psu.edu/viewdoc/summary?doi=",
  URL =          "http://serg.aau.at/pub/MartinPinzger/Publications/Ratzinger2007-quality.pdf",
  DOI =          "doi:10.1109/WCRE.2007.39",
  abstract =     "Defect density and defect prediction are essential for
                 efficient resource allocation in software evolution. In
                 an empirical study we applied data mining techniques
                 for value series based on evolution attributes such as
                 number of authors, commit messages, lines of code, bug
                 fix count, etc. Daily data points of these evolution
                 attributes were captured over a period of two months to
                 predict the defects in the subsequent two months in a
                 project. For that, we developed models using genetic
                 programming and linear regression to accurately predict
                 software defects. In our study, we investigated the
                 data of three independent projects, two open source and
                 one commercial software system. The results show that
                 by using series of these attributes we obtain models
                 with high correlation coefficients (between 0.716 and
                 0.946). Further, we argue that prediction models based
                 on series of a single variable are sometimes superior
                 to the model including all attributes: in contrast to
                 other studies that resulted in size or complexity
                 measures as predictors, we have identified the number
                 of authors and the number of commit messages to
                 versioning systems as excellent predictors of defect
  notes =        "also known as \cite{4400154}",

Genetic Programming entries for Jacek Ratzinger Harald Gall Martin Pinzger