Moussa, Rebecca; Guizzo, Giovani; Sarro, Federica; (2023) MEG: Multi-objective Ensemble Generation for Software Defect Prediction (HOP GECCO'23). In: Proceedings of the GECCO '23 Companion: Proceedings of the Companion Conference on Genetic and Evolutionary Computation. (pp. pp. 37-38). ACM

Preview

Text moussa-gecco-hop.pdf - Accepted Version Download (914kB) | Preview

Abstract

This Hot-off-the-Press abstract aims at disseminating our recent work titled “MEG: Multi-objective Ensemble Generation for Software Defect Prediction” published in the proceedings of the 16th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM) [4]. We believe this work is of interest for the GECCO community as it proposes a novel way to automatically generate ensemble machine learning models leveraging the power of evolutionary computation: MEG introduces the concept of whole-ensemble generation as opposed to the well known Pareto-ensemble generation. While we evaluate the effectiveness of MEG for Software Defect Prediction in our work, MEG can be applied to any classification or regression problem and we invite both researchers and practitioners to further explore its effectiveness for other application domains. To this end, we have made MEG’s source code publicly available.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item