Bachelor and Master Theses

Title: Performance evaluation of multi-objective search based techniques for software fault prediction
Subject: Computer Science
Level: Advanced
Description: The goal of software fault prediction is to help software engineers focus testing and refactoring activities on fault-prone code, thus improving software quality. There are many software fault prediction studies in software engineering literature, see e.g. recent literature review papers on the topic [1] [2]. However there is very little evidence available on the performance of fault prediction models in the presence of conflicting objectives (e.g., prediction quality vs. cost). Few studies have targeted this problem, see e.g. [3] [4] however there are still many other competing objectives that need to be balanced. Search-based techniques are well-suited to deal with conflicting objectives, see [5]. The goal with this thesis is to evaluate a number of multi-objective search-based techniques in terms of prediction performance and to present a rich description of pros and cons of using such techniques.

Expected outcomes:
- Performance evaluation of a number of multi-objective search-based techniques for software fault prediction.
- Discussion of pros and cons of using such techniques.

References:

[1] Hall, T.; Beecham, S.; Bowes, D.; Gray, D.; Counsell, S., "A Systematic Literature Review on Fault Prediction Performance in Software Engineering," in Software Engineering, IEEE Transactions on , vol.38, no.6, pp.1276-1304, Nov.-Dec. 2012.

[2] Ruchika Malhotra, A systematic review of machine learning techniques for software fault prediction, Applied Soft Computing, Volume 27, February 2015, Pages 504-518, ISSN 1568-4946, http://dx.doi.org/10.1016/j.asoc.2014.11.023.

[3] Gerardo Canfora, Andrea De Lucia, Massimiliano Di Penta, Rocco Oliveto, Annibale Panichella, Sebastiano Panichella:
Defect prediction as a multiobjective optimization problem. Softw. Test., Verif. Reliab. 25(4): 426-459 (2015)

[4] André B. de Carvalho, Aurora Pozo, Silvia Regina Vergilio, A symbolic fault-prediction model based on multiobjective particle swarm optimization, Journal of Systems and Software, Volume 83, Issue 5, May 2010, Pages 868-882, ISSN 0164-1212, http://dx.doi.org/10.1016/j.jss.2009.12.023.
(http://www.sciencedirect.com/science/article/pii/S0164121209003367)

[5] Mark Harman. 2010. The relationship between search based software engineering and predictive modeling. In Proceedings of the 6th International Conference on Predictive Models in Software Engineering (PROMISE '10). ACM, New York, NY, USA, , Article 1 , 13 pages. DOI=http://dx.doi.org/10.1145/1868328.1868330
Proposed: 2016-10-31
Prerequisites: 1) Programming skills 2) Research methods 3) Optimization
IDT supervisor: Wasif Afzal
wasif.afzal@mdh.se,
Examinator:

,

Rapport och bilagor

Size

Senaste uppdatering


  • Mälardalen University |
  • Box 883 |
  • 721 23 Västerås/Eskilstuna |
  • 021-101300, 016-153600 |
  • webmaster |
  • Latest update: 2017.08.23