• DocumentCode
    3513237
  • Title

    An Early Prediction Method of Software Reliability Based on Support Vector Machine

  • Author

    Li, Xingguo ; Li, Xiaofeng ; Shu, Yanhua

  • Author_Institution
    Dept. of Inf. Manage., Hefei Univ. of Technol., Hefei
  • fYear
    2007
  • fDate
    21-25 Sept. 2007
  • Firstpage
    6075
  • Lastpage
    6078
  • Abstract
    With rapid development of information technology, we are more and more dependent on computer systems. Computer software is the primary factor among all that can result in computer system errors. Therefore, how to design and develop reliable software, and how to ensure its quality is an urgent task. Early prediction of software reliability can enable developers to obtain general ideas about software reliability before testing it. This is critical for further development, testing and quality control of the software. This paper summarizes the status of early prediction methods for software reliability in both China and in other countries. We have introduced the support vector machine (SVM) theory into the study on early prediction of software reliability, and advanced the software reliability early prediction model based on SVM. Simulation shows that comparing with the classic models, this model is more precise in its prediction capacity, and has better capacity in generalization, and is less dependent on the sample size.
  • Keywords
    program testing; software quality; software reliability; support vector machines; computer software; computer system errors; early prediction method; reliable software; software quality control; software reliability; software testing; support vector machine; Computer errors; Information technology; Prediction methods; Predictive models; Quality control; Software quality; Software reliability; Software systems; Software testing; Support vector machines;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wireless Communications, Networking and Mobile Computing, 2007. WiCom 2007. International Conference on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-1311-9
  • Type

    conf

  • DOI
    10.1109/WICOM.2007.1490
  • Filename
    4341265