Learning cost sensitive binary classification rules accounting for uncertain and unequal misclassification costs

Arbeitspapier

This paper proposes cost sensitive criteria for constructing classification rules by supervised learning methods. Reinterpreting established loss functions and considering those introduced by Buja, Stuetzle, et al. (2005) and Hand (2009), we identify criteria reflecting different degrees of information about misclassification costs. To adapt classification methodology to practical cost considerations, we suggest the use of these criteria for different model selection approaches in supervised learning. In addition, we investigate the effects of cost sensitive adaptations in CART and boosting and conclude that adaptations are more promising in the selection rather than in the estimation step.

Angaben zum Objekt

• Sprache: Englisch

Verweise und Beziehungen

• Erschienen in: Series: IWQW Discussion Papers ; No. 01/2014

Klassifikation und Themen

• Klassifikation: Wirtschaft
• Thema: unequal misclassification costs ; proper scoring rules ; AUC ; boosting ; CART ; model selection ; pruning ; early stopping

Beteiligte, Orts- und Zeitangaben

• Ereignis: Geistige Schöpfung
• (wer): Rybizki, Lydia
• Ereignis: Veröffentlichung
• (wer): Friedrich-Alexander-Universität Erlangen-Nürnberg, Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung (IWQW)
• (wo): Nürnberg
• (wann): 2014

Weitere Informationen

• Handle: http://hdl.handle.net/10419/94192
• Letzte Aktualisierung: 10.03.2025, 11:41 MEZ