Experimental Methods in Computer Science

Recommended Prerequisites

Statistics.

Teaching Methods

The classes contain theoretical exposition about experimental methodologies and their application in the context of computer science in terms of case studies. The assessment consists of designing and analysing experiments from computational systems under several scenarios.

Learning Outcomes

To conduct empirical assessment of computational systems that arise in several topics of computer science, such as communication networks, algorithms and artificial intelligence. Acquiring competencies in

analysis and synthesis, organization and planning.

Work Placement(s)

No

Syllabus

1) The experimental method; experimental scenarios in computer science

2) Exploratory data analysis

3) Measurements and benchmarking

4) Data analysis: distributions, correlation, linear regression

5) Experimental design: factors and interactions

6) Qualitative methodology: interviews and forms, content analysis.

Head Lecturer(s)

Luís Filipe dos Santos Coelho Paquete

Assessment Methods

Assessment
Project: 100.0%

Bibliography

               1) T. Bartz-Beielstein, M. Chiarandini, L. Paquete, M. Preuss, Experimental Methods for the Analysis of Optimization Algorithms, Springer, 2010.

                2) P. Cohen, Empirical Methods for Artificial Intelligence, MIT Press, 1995

                3) R. Jain, The Art of Computer Systems Performance Analysis, Wiley 1991.

                4) D.J. Lilja, Measuring Computer Performance, Cambridge University Press, 2000

                5) J. Lazar, J. Feng, H. Hochheiser, Research Methods in Human-Computer Interaction, (Chapter11 - Analyzing qualitative data), John Wiley and Sons, 2010.

                6) C.C. McGeoch, A Guide to Experimental Algorithmics, Cambrigde University Press, 2012