Novelty Discovery with Ensemble Kernel Minimum Enclosing Balls

We introduce the idea of utilizing ensembles of Kernel Minimum Enclosing Balls to detect novel datapoints. To this end, we propose a novelty scoring methodology that is based on combining the outcomes of the corresponding characteristic functions of the fitted balls. We empirically evaluate our model by presenting experiments on synthetic as well as real world datasets.

  • Published in:
    LION 2020: Learning and Intelligent Optimization Learning and Intelligent Optimization (LION)
  • Type:
    Inproceedings
  • Authors:
    R. Sifa, C. Bauckhage
  • Year:
    2020

Citation information

R. Sifa, C. Bauckhage: Novelty Discovery with Ensemble Kernel Minimum Enclosing Balls, Learning and Intelligent Optimization (LION), LION 2020: Learning and Intelligent Optimization, 2020, https://doi.org/10.1007/978-3-030-53552-0_37, Sifa.Bauckhage.2020,