Simulated annealing for improving software quality prediction
Proceedings of the 8th annual conference on Genetic and evolutionary computation
A novel composite model approach to improve software quality prediction
Information and Software Technology
The relationship between search based software engineering and predictive modeling
Proceedings of the 6th International Conference on Predictive Models in Software Engineering
Combining techniques for software quality classification: An integrated decision network approach
Expert Systems with Applications: An International Journal
Survey: A survey on search-based software design
Computer Science Review
Search-based software engineering: Trends, techniques and applications
ACM Computing Surveys (CSUR)
| Hi-index | 0.00 |
During the past ten years, a large number of qualitymodels have been proposed in the literature. In general,the goal of these models is to predict a quality factor startingfrom a set of direct measures. The lack of data behindthese models makes it hard to generalize, to cross-validate,and to reuse existing models. As a consequence, for a company,selecting an appropriate quality model is a difficult,non-trivial decision. In this paper, we propose a general approachand a particular solution to this problem. The mainidea is to combine and adapt existing models (experts) insuch way that the combined model works well on the particularsystem or in the particular type of organization. In ourparticular solution, the experts are assumed to be decisiontree or rule-based classifiers and the combination is doneby a genetic algorithm. The result is a white-box model: foreach software component, not only the model gives the predictionof the software quality factor, but it also provides theexpert that was used to obtain the prediction. Test results indicatethat the proposed model performs significantly betterthan individual experts in the pool.