semanticscholar.org

On the evaluation of unsupervised outlier detection: measures, datasets, and an empirical study | Semantic Scholar

@article{Campos2016OnTE,
  title={On the evaluation of unsupervised outlier detection: measures, datasets, and an empirical study},
  author={Guilherme Oliveira Campos and Arthur Zimek and J{\"o}rg Sander and Ricardo J. G. B. Campello and Barbora Micenkov{\'a} and Erich Schubert and Ira Assent and Michael E. Houle},
  journal={Data Mining and Knowledge Discovery},
  year={2016},
  volume={30},
  pages={891-927},
  url={https://api.semanticscholar.org/CorpusID:1952214}
}

An extensive experimental study on the performance of a representative set of standard k nearest neighborhood-based methods for unsupervised outlier detection, across a wide variety of datasets prepared for this purpose, and provides a characterization of the datasets themselves.

634 Citations

AutoOD: Automatic Outlier Detection

This work proposes AutoOD, which uses the existing unsupervised detection techniques to automatically produce high quality outliers without any human tuning, and which consistently outperforms the best un supervised outlier detector selected from hundreds of detectors.

67 References

Outlier detection by active learning

This paper presents a novel approach to outlier detection based on classification, which is superior to other methods based on the same reduction to classification, but using standard classification methods, and shows that it is competitive to the state-of-the-art outlier Detection methods in the literature.