Publication detail

Area under the ROC Curve by Bubble-Sort Approach (BSA)

HONZÍK, P.

Original Title

Area under the ROC Curve by Bubble-Sort Approach (BSA)

English Title

Area under the ROC Curve by Bubble-Sort Approach (BSA)

Type

conference paper

Language

en

Original Abstract

A new approach to area under ROC curve (AUC) evaluation is introduced and compared with the current methods. The main idea is based on the Bubble-Sort method. It advantages from the approach to the qualitative dependent variable which is used as the ordinal (and not nominal) variable in comparison to the classical approach. For binary output data the algorithm reaches the same complexity and values as the other methods. For multi-class classification the complexity differs from the classical approaches and the BSA values of AUC don't fail in the special cases as the current methods do.

English abstract

A new approach to area under ROC curve (AUC) evaluation is introduced and compared with the current methods. The main idea is based on the Bubble-Sort method. It advantages from the approach to the qualitative dependent variable which is used as the ordinal (and not nominal) variable in comparison to the classical approach. For binary output data the algorithm reaches the same complexity and values as the other methods. For multi-class classification the complexity differs from the classical approaches and the BSA values of AUC don't fail in the special cases as the current methods do.

RIV year

2005

Released

13.03.2005

Publisher

WSEAS

Location

Praha

ISBN

960-8457-12-2

Book

Automatic Control, Modeling and Simulation (ACMOS'05)

Pages from

494

Pages to

499

Pages count

6

URL

BibTex


@inproceedings{BUT17493,
  author="Petr {Honzík}",
  title="Area under the ROC Curve by Bubble-Sort Approach (BSA)",
  annote="A new approach to area under ROC curve (AUC) evaluation is introduced and compared with the current methods. The main idea is based on the Bubble-Sort method. It advantages from the approach to the qualitative dependent variable which is used as the ordinal (and not nominal) variable in comparison to the classical approach. For binary output data the algorithm reaches the same complexity and values as the other methods. For multi-class classification the complexity differs from the classical approaches and the BSA values of AUC don't fail in the special cases as the current methods do.",
  address="WSEAS",
  booktitle="Automatic Control, Modeling and Simulation (ACMOS'05)",
  chapter="17493",
  institution="WSEAS",
  year="2005",
  month="march",
  pages="494",
  publisher="WSEAS",
  type="conference paper"
}