In this keynote I introduce the use of Predictive Analytics for Software Engineering (SE) and then focus on the use of search-based heuristics to tackle long-standing SE prediction problems including (but not limited to) software development effort estimation and software defect prediction. I review recent research in Search-Based Predictive Modelling for SE in order to assess the maturity of the field and point out promising research directions. I conclude my keynote by discussing best practices for a rigorous and realistic empirical evaluation of search-based predictive models, a condicio sine qua non to facilitate the adoption of prediction models in software industry practices.
CITATION STYLE
Sarro, F. (2019). Search-Based Predictive Modelling for Software Engineering: How Far Have We Gone? In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11664 LNCS, pp. 3–7). Springer Verlag. https://doi.org/10.1007/978-3-030-27455-9_1
Mendeley helps you to discover research relevant for your work.