狠狠撸s experimentation softengineering
?
1 like?149 views
The document introduces experimentation in software engineering, discussing the experiment process, examples of experiments comparing C and C++, and conclusions. It describes the stages of an experiment including definition and planning, operation, analysis and interpretation. Examples provided compare productivity, defects, and other measures between C and C++. The conclusion recommends the document as an introduction to using experimentation for software engineering evaluation.
狠狠撸s experimentation softengineering
- 1. Experimentation in Software Engineering: A Introduction Claes Wohlim, et al. Lund University Sweden Presented by: Diogo de Jesus Pina Marcos Amar??s Gonz?alez University of S?ao Paulo Institute of Mathematics and Statistics Department of Computer Science November, 2014 {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 1 / 17
- 2. Timeline 1 Introduction 2 Experiment Process De?nition, Planning and Operation Analysis and Interpretation 3 Examples 4 Conclusions {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 2 / 17
- 3. Introduction Introduction Software Engineering Context Science and Software Engineering {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 3 / 17
- 4. Experiment Process Empirical Strategies Factor Survey Case Study Experiment Execution control No No Yes Measurement control No Yes Yes Investigation cost Low Medium High Ease of replication High Low High {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 4 / 17
- 5. Experiment Process Measurement Basic Concepts Measurements in Software Engineering Measurements in Practice {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 5 / 17
- 6. Experiment Process Systematic Literature Reviews Planning Conducting Mapping Example {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 6 / 17
- 7. Experiment Process Case Studies Software Engineering Context Design and Planning Preparation and Collection of Data Data Analysis Reporting {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 7 / 17
- 8. Experiment Process Experiment Process {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 8 / 17
- 9. Experiment Process Scoping {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 9 / 17
- 10. Experiment Process De?nition, Planning and Operation Planing {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 10 / 17
- 11. Experiment Process De?nition, Planning and Operation Operation {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 11 / 17
- 12. Experiment Process Analysis and Interpretation Analysis and Interpretation {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 12 / 17
- 13. Experiment Process Analysis and Interpretation Presentation and Package {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 13 / 17
- 14. Examples Examples Process Experimentation Personal Software Process is a individual process for a systematic approach to software development. The process includes, for example, measurement, estimation, planning, and tracking. Software Development Training Course De?nition: Productivity, defect density, experience in C. Planning: prod(CS) = prod(EE), LOC. Operation Analysis and Interpretation {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 14 / 17
- 15. Examples Examples C versus C++ Personal Software Process (C versus C++) Data: Ten di?erent pograms. From: Lund97 Msc. Student and Link96 Ph.D. Students. Experimentation is made with respect to eight di?erent measures. 1 Total time 2 Planning time 3 Code to test 4 Code to test relative 1 Total quality test 2 Total defects 3 Code defects 4 Review e?ciency The results are not statistically signi?cant, although they indicate that more defects are introduced if C++ is use than if C is used. {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 15 / 17
- 16. Conclusions Conclusions The purpose of this book is to introduce to experimentation and ex- perimental evaluation with focus of software engineering, as a means for evaluating methods, techniques, processes, products, etc. It was interesting Technical reading Easy reading Useful We recommend it. {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 16 / 17
- 17. {diogojp, amaris}@ime.usp.br (IME - USP) Experimentation in S.E. November, 2014 17 / 17