---
_id: '2109'
abstract:
- lang: eng
  text: In multinomial classification, reduction techniques are commonly used to decompose
    the original learning problem into several simpler problems. For example, by recursively
    bisecting the original set of classes, so-called nested dichotomies define a set
    of binary classification problems that are organized in the structure of a binary
    tree. In contrast to the existing one-shot heuristics for constructing nested
    dichotomies and motivated by recent work on algorithm configuration, we propose
    a genetic algorithm for optimizing the structure of such dichotomies. A key component
    of this approach is the proposed genetic representation that facilitates the application
    of standard genetic operators, while still supporting the exchange of partial
    solutions under recombination. We evaluate the approach in an extensive experimental
    study, showing that it yields classifiers with superior generalization performance.
author:
- first_name: Marcel Dominik
  full_name: Wever, Marcel Dominik
  id: '33176'
  last_name: Wever
  orcid: ' https://orcid.org/0000-0001-9782-6818'
- first_name: Felix
  full_name: Mohr, Felix
  last_name: Mohr
- first_name: Eyke
  full_name: Hüllermeier, Eyke
  id: '48129'
  last_name: Hüllermeier
citation:
  ama: 'Wever MD, Mohr F, Hüllermeier E. Ensembles of Evolved Nested Dichotomies for
    Classification. In: <i>Proceedings of the Genetic and Evolutionary Computation
    Conference, GECCO 2018, Kyoto, Japan, July 15-19, 2018</i>. Kyoto, Japan: ACM;
    2018. doi:<a href="https://doi.org/10.1145/3205455.3205562">10.1145/3205455.3205562</a>'
  apa: 'Wever, M. D., Mohr, F., &#38; Hüllermeier, E. (2018). Ensembles of Evolved
    Nested Dichotomies for Classification. In <i>Proceedings of the Genetic and Evolutionary
    Computation Conference, GECCO 2018, Kyoto, Japan, July 15-19, 2018</i>. Kyoto,
    Japan: ACM. <a href="https://doi.org/10.1145/3205455.3205562">https://doi.org/10.1145/3205455.3205562</a>'
  bibtex: '@inproceedings{Wever_Mohr_Hüllermeier_2018, place={Kyoto, Japan}, title={Ensembles
    of Evolved Nested Dichotomies for Classification}, DOI={<a href="https://doi.org/10.1145/3205455.3205562">10.1145/3205455.3205562</a>},
    booktitle={Proceedings of the Genetic and Evolutionary Computation Conference,
    GECCO 2018, Kyoto, Japan, July 15-19, 2018}, publisher={ACM}, author={Wever, Marcel
    Dominik and Mohr, Felix and Hüllermeier, Eyke}, year={2018} }'
  chicago: 'Wever, Marcel Dominik, Felix Mohr, and Eyke Hüllermeier. “Ensembles of
    Evolved Nested Dichotomies for Classification.” In <i>Proceedings of the Genetic
    and Evolutionary Computation Conference, GECCO 2018, Kyoto, Japan, July 15-19,
    2018</i>. Kyoto, Japan: ACM, 2018. <a href="https://doi.org/10.1145/3205455.3205562">https://doi.org/10.1145/3205455.3205562</a>.'
  ieee: M. D. Wever, F. Mohr, and E. Hüllermeier, “Ensembles of Evolved Nested Dichotomies
    for Classification,” in <i>Proceedings of the Genetic and Evolutionary Computation
    Conference, GECCO 2018, Kyoto, Japan, July 15-19, 2018</i>, Kyoto, Japan, 2018.
  mla: Wever, Marcel Dominik, et al. “Ensembles of Evolved Nested Dichotomies for
    Classification.” <i>Proceedings of the Genetic and Evolutionary Computation Conference,
    GECCO 2018, Kyoto, Japan, July 15-19, 2018</i>, ACM, 2018, doi:<a href="https://doi.org/10.1145/3205455.3205562">10.1145/3205455.3205562</a>.
  short: 'M.D. Wever, F. Mohr, E. Hüllermeier, in: Proceedings of the Genetic and
    Evolutionary Computation Conference, GECCO 2018, Kyoto, Japan, July 15-19, 2018,
    ACM, Kyoto, Japan, 2018.'
conference:
  end_date: 2018-07-19
  location: Kyoto, Japan
  name: GECCO 2018
  start_date: 2018-07-15
date_created: 2018-03-31T13:51:23Z
date_updated: 2022-01-06T06:54:45Z
ddc:
- '000'
department:
- _id: '355'
doi: 10.1145/3205455.3205562
file:
- access_level: closed
  content_type: application/pdf
  creator: ups
  date_created: 2018-11-02T14:33:54Z
  date_updated: 2018-11-02T14:33:54Z
  file_id: '5275'
  file_name: p561-wever.pdf
  file_size: 875404
  relation: main_file
  success: 1
file_date_updated: 2018-11-02T14:33:54Z
has_accepted_license: '1'
keyword:
- Classification
- Hierarchical Decomposition
- Indirect Encoding
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://dl.acm.org/citation.cfm?doid=3205455.3205562
oa: '1'
place: Kyoto, Japan
project:
- _id: '1'
  name: SFB 901
- _id: '3'
  name: SFB 901 - Project Area B
- _id: '10'
  name: SFB 901 - Subproject B2
- _id: '52'
  name: Computing Resources Provided by the Paderborn Center for Parallel Computing
publication: Proceedings of the Genetic and Evolutionary Computation Conference, GECCO
  2018, Kyoto, Japan, July 15-19, 2018
publication_status: published
publisher: ACM
status: public
title: Ensembles of Evolved Nested Dichotomies for Classification
type: conference
user_id: '33176'
year: '2018'
...