On Usefulness of Outlier Elimination in Classification Tasks

HETLEROVIĆ, Dušan, Lubomír POPELÍNSKÝ, P. BRAZDIL, C. SOARES and F. FREAITAS. On Usefulness of Outlier Elimination in Classification Tasks. Online. In Tassadit Bouadi, Élisa Fromont, Eyke Hüllermeier. International Symposium on Intelligent Data Analysis 2022. Rennes: Springer, 2022, p. 143-156. ISBN 978-3-031-01332-4. Available from: https://dx.doi.org/10.1007/978-3-031-01333-1_12.

Basic information

• Original name: On Usefulness of Outlier Elimination in Classification Tasks
• Name in Czech: On Usefulness of Outlier Elimination in Classification Tasks
• Authors: HETLEROVIĆ, Dušan (703 Slovakia, belonging to the institution), Lubomír POPELÍNSKÝ (203 Czech Republic, belonging to the institution), P. BRAZDIL, C. SOARES and F. FREAITAS.
• Edition: Rennes, International Symposium on Intelligent Data Analysis 2022, p. 143-156, 14 pp. 2022.
• Publisher: Springer

Other information

• Original language: English
• Type of outcome: Proceedings paper
• Field of Study: 10201 Computer sciences, information science, bioinformatics
• Confidentiality degree: is not subject to a state or trade secret
• Publication form: electronic version available online
• Impact factor: Impact factor: 0.402 in 2005
• RIV identification code: RIV/00216224:14330/22:00126186
• Organization unit: Faculty of Informatics
• ISBN: 978-3-031-01332-4
• ISSN: 0302-9743
• Doi: http://dx.doi.org/10.1007/978-3-031-01333-1_12
• UT WoS: 000937256100012
• Keywords (in Czech): Outlier elimination; Metalearning; Average ranking; Reduction of portfolios
• Keywords in English: Outlier elimination; Metalearning; Average ranking; Reduction of portfolios
• Tags: firank_A
• Tags: International impact, Reviewed

Abstract

Although outlier detection/elimination has been studied before, few comprehensive studies exist on when exactly this technique would be useful as preprocessing in classification tasks. The objective of our study is to fill in this gap. We have performed experiments with 12 various outlier elimination methods and 10 classification algorithms on 50 different datasets. The results were then processed by the proposed reduction method, whose aim is identify the most useful workflows for a given set of tasks (datasets). The reduction method has identified that just three OEMs that are generally useful for the given set of tasks. We have shown that the inclusion of these OEMs is indeed useful, as it leads to lower loss in accuracy and the difference is quite significant (0.5\%) on average.

Abstract (in Czech)

Although outlier detection/elimination has been studied before, few comprehensive studies exist on when exactly this technique would be useful as preprocessing in classification tasks. The objective of our study is to fill in this gap. We have performed experiments with 12 various outlier elimination methods and 10 classification algorithms on 50 different datasets. The results were then processed by the proposed reduction method, whose aim is identify the most useful workflows for a given set of tasks (datasets). The reduction method has identified that just three OEMs that are generally useful for the given set of tasks. We have shown that the inclusion of these OEMs is indeed useful, as it leads to lower loss in accuracy and the difference is quite significant (0.5\%) on average.