Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A Comparison and Integration of Capture-Recapture Models and the Detection Profile Method

: Briand, L.C.; Emam, K. el; Freimut, B.

Fulltext urn:nbn:de:0011-px-449717 (674 KByte PDF)
MD5 Fingerprint: d10278a9916ace9915f0d07a7ebfd12a
Created on: 09.08.2000

Kaiserslautern, 1998, 15 pp. : Ill., Lit.
IESE-Report, 025.98/E
Reportnr.: IESE-Report 025.98/E
Report, Electronic Publication
Fraunhofer IESE ()
capture-recapture model; defect content estimation; software inspection

In order to control inspections, the number of remaining defects in software artifacts after their inspection should be estimated. This would allow, for example, deciding whether a reinspection of supposedly faulty artefacts is necessary. Several studies in software engineering have considered capture-recapture models for performing such estimations. These models were initially developed for estimating animal abundance in wildlife research. In addition to these models, researchers in software engineering have recently proposed a procedure, namely the Detection Profile Method (DPM), that makes less restrictive assumptions than some capture-recapture models and that show promise in terms of estimation accuracy. In this study, we investigate the combination of DPM with capture-recapture models to address a practical difficulty when applying capture-recapture models alone: extreme under/over estimation. The existence of such extreme outliers provided by capture-recapture models can discour age their use because their consequences in terms of wasted effort of defect slippage can be substantial, and therefore it is not clear whether a particular estimate can be trusted. We identify a hybrid approach, using both capture-recapture models and DPM, and evaluate it using actual inspection data. Our results indicate that this hybrid approach has the same accuracy as capture-recapture models alone and DPM alone, and most importantly does not exhibit extreme over/under estimation. This new approach can be used in practice with a high degree of confidence since its estimates are not likely to exhibit extreme estimation error.