Proposal

Florida Atlantic University
Empirical Software Engineering Laboratory

Proposal for Full-Day Tutorial on
Building a Corporate Metrics Program for High Quality Software

Taghi M. Khoshgoftaar Edward B. Allen
taghi@cse.fau.edu allene@cse.fau.edu
Empirical Software Engineering Laboratory
Dept. of Computer Science and Engineering
Florida Atlantic University
Boca Raton, FL 33431 USA

Abstract
High software quality is the goal of a software metrics program. Sofware product and process metrics are the raw material that you can use to predict and evaluate software quality. Which software metrics will make sense for your development process? Which modules are likely to be fault-prone in the future? What parts of a system will have high code churn? This tutorial teaches you how to build a software metrics program that will incorporate the best ideas from academia and industry, including how to select software product and quality metrics, how to formulate software quality models, and how to validate a model for your development process. We will draw on our consulting experience with industry. Practical examples, based on data from full size industrial software systems, will illustrate the techniques discussed.

Keywords: software quality, software engineering, software metrics, measurement,
modeling, validation, technology transfer

1 Overview
Sofware product and process metrics are the raw material that you can use to predict and evaluate software quality. This tutorial teaches how to build a software metrics program that will make a positive contribution to software quality. The tutorial is intended for software development practitioners with interests in the technical and managerial aspects of software quality modeling. Practical examples, based on data from full size industrial software systems, will illustrate the techniques discussed.

2 Background
Dr. Khoshgoftaar and Dr. Allen have presented versions of this tutorial to industry clients and at the 1997 Conference on Computer Assurance (COMPASS'97).

3 Objective
The objective of the tutorial is to convey an appreciation of the many issues involved in a successful and useful software metrics program. Especially important are software quality modeling issues and the practical use of modeling results.

4 Audience
The tutorial is intended for software development practitioners with interests in the technical and managerial aspects of software quality modeling. The in-depth view of software metrics collection and modeling will be of special interest to the technical audience. Discussions on the use of metrics, application of models, and technology transfer from academia will be of general interest.

5 Prerequisites
General knowledge of software engineering is beneficial. Introductory knowledge of probability concepts is beneficial for software quality modeling topics.

6 Presenters
Taghi M. Khoshgoftaar received an MS in applied mathemetics from Massachusetts Institute of Technology, an MS in computer science from North Carolina State University, and a PhD in statistics from Virginia Polytechnic Institute and State University. He is a professor of the Dept. of Computer Science and Engineering, Florida Atlantic University. He is also the Director of the Empirical Software Engineering Laboratory, established through a grant from the National Science Foundation. His research interests are in software engineering, software complexity metrics and measurements, software reliability and quality engineering, neural networks applications, computer performance evaluation, multimedia systems, and statistical modeling. He has published more than 100 refereed papers in these areas. He is a member of the Association for Computing Machinery, the American Statistical Association, and the IEEE (Computer Society and Reliability Society). He has served on technical program committees of various international conferences, symposia, and workshops. He is North American editor of the Software Quality Journal, and is on the editorial board of the Journal of Multimedia Tools and Applications.

Edward B. Allen received the B.S. degree in engineering from Brown University, Providence, RI, USA, in 1971, the M.S. degree in systems engineering from the Univer sity of Pennsylvania, Philadelphia, PA, USA, in 1973, and the Ph.D. degree in computer science from Florida Atlanatic University, Boca Raton, FL, USA, in 1995. He is currently a Research Associate and an adjunct professor in the Department of Computer Science and Engineering at Florida Atlantic University. He began his career as a programmer with the U.S. Army. From 1974 to 1983, he performed systems engineering and software engineering on military systems, first for Planning Research Corp. and then for Sperry Corp. From 1983 to 1992, he developed corporate data processing systems for Glenbeigh, Inc., a specialty health care company. His research interests include software measurement, software process modeling, and software quality. He is a member of the IEEE Computer Society and the Association for Computing Machinery.

7 Outline
1 Introduction
2 Context
3 Using Metrics
4 Using Software Quality Models
5 Preparing Data Sets
6 Building Software Quality Models
6.1 Selecting Software Metrics
6.2 Principal Components Analysis
6.3 Quantity Models
6.4 Classification Models
6.5 Process Metrics
7 Validating Software Quality Models
7.1 Quantity Models
7.2 Classification Models
8 From Lab to Practice

Last update: 24 Aug 98 eba
Comments about this web page?
Contact webmaster@cse.fau.edu
URL: http://www.cse.fau.edu/research/ESEL/tutorial.htm