An investigation on the use of machine learned models for estimating correction costs

Author

De Almeida, Mauricio A. ; Lounis, Hakim ; Melo, Walcélio L.

Author_Institution

SITE, Ottawa Univ., Ont., Canada

fYear

1998

fDate

19-25 Apr 1998

Firstpage

473

Lastpage

476

Abstract

We present the results of an empirical study in which we have investigated machine learning (ML) algorithms with regard to their capabilities to accurately assess the correctability of faulty software components. Three different families of algorithms have been analyzed. We have used (1) fault data collected on corrective maintenance activities for the Generalized Support Software reuse asset library located at the Flight Dynamics Division of NASA´s GSFC and (2) product measures extracted directly from the faulty components of this library

Keywords

learning (artificial intelligence); software cost estimation; software development management; software libraries; software maintenance; software reusability; Flight Dynamics Division; Generalized Support Software; NASA; coding guidelines; correctability; correction costs estimation; corrective maintenance activities; empirical study; fault data; faulty software components; machine learning algorithms; predictive software quality model building; product measures; reuse asset library; Costs; Educational institutions; Error correction; Machine learning; Machine learning algorithms; Predictive models; Software algorithms; Software libraries; Software maintenance; Software quality;

fLanguage

English

Publisher

ieee

Conference_Titel

Software Engineering, 1998. Proceedings of the 1998 International Conference on

Conference_Location

Kyoto

ISSN

0270-5257

Print_ISBN

0-8186-8368-6

Type

conf

DOI

10.1109/ICSE.1998.671609

Filename

671609