Software Quality Imputation in the Presence of Noisy Data

Author

Khoshgoftaar, Taghi M. ; Folleco, Andres ; Van Hulse, Jason ; Bullard, Lofton

Author_Institution

Dept. of Comput. Sci. & Eng., Florida Atlantic Univ., Boca Raton, FL

fYear

2006

fDate

16-18 Sept. 2006

Firstpage

484

Lastpage

489

Abstract

The detrimental effects of noise in a dependent variable on the accuracy of software quality imputation techniques were studied. The imputation techniques used in this work were Bayesian multiple imputation, mean imputation, instance-based learning, regression imputation, and the REPTree decision tree. These techniques were used to obtain software quality imputations for a large military command, control, and communications system dataset (CCCS). The underlying quality of data was a significant factor affecting the accuracy of the imputation techniques. Multiple imputation and regression imputation were top performers, while mean imputation was ineffective

Keywords

belief networks; decision trees; regression analysis; software quality; Bayesian multiple imputation; REPTree decision tree; instance-based learning; mean imputation; military command-control-communication system; regression imputation; software measurement; software quality imputation; Bayesian methods; Computer science; Data engineering; Data mining; Decision trees; Laboratories; Regression tree analysis; Software engineering; Software measurement; Software quality; Bayesian multiple imputation; inherent/simulated noise; missing data; software measurement; software quality;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Reuse and Integration, 2006 IEEE International Conference on

Conference_Location

Waikoloa Village, HI

Print_ISBN

0-7803-9788-6

Type

conf

DOI

10.1109/IRI.2006.252462

Filename

4018539