Author Type

Graduate Student

Date of Award

2002

Document Type

Thesis

Publication Status

Version of Record

Submission Date

October 2026

Thesis/Dissertation Advisor [Chair]

Khoshgoftaar, Taghi M.

Abstract

Software managers are under pressure to deliver reliable and high quality software, within a limited time and budget. To achieve this goal, they can be aided by different modeling techniques that allow them to predict the quality of software, so that the improvement efforts can be directed to software modules that are more likely to be fault-prone. Also, different projects have different resource availability constraints, and being able to select a model that is suitable for a specific resource constraint allows software managers to direct enhancement techniques more effectively and efficiently. In our study, we use Rule-Based Modeling ( RBM) to predict the likelihood of a module being fault-prone and the Modified Expected Cost of Misclassification (MECM ) measure to select the models that are suitable, in the context of the given resource constraints. This empirical study validates MECM as a measure to select an appropriate RBM model.

Extent

169 p.

Comments

College of Engineering and Computer Science|FAU Electronic Theses and Dissertations Collection|Thesis (M.S.)--Florida Atlantic University, 2002.|College of Engineering and Computer Science

Identifier

FADT12968

Format

pdf

Download

Share

COinS