Confidence-based ensemble modeling in medical data mining

Lukas Kammerer, Michael Affenzeller

Abstract

A recent approach for improving the accuracy of ensemble models is confidence-based modeling. Thereby, confidence measures, which indicate an ensemble prediction's reliability, are used for identifying unreliable predictions in order to improve a model's accuracy among reliable predictions. However, despite promising results in previous work, no comparable results for public benchmark data sets have been published yet. This paper applies confidence-based modeling with GP-based symbolic binary classification ensembles on a set of medical benchmark problems to make statements about the concept's general applicability. Moreover, extensions for multiclass classification problems are proposed.
BibTeX
@inproceedings{kammerer2018confidence,
  author = {Lukas Kammerer and Michael Affenzeller},
  title = {{Confidence-based ensemble modeling in medical data mining}},
  booktitle = {GECCO (Companion)},
  year = {2018},
  doi = {10.1145/3205651.3205722},
}

Loading related content...

Figure