Using the receiver operating characteristic to assess the performance of neural classifiers

Thomas J. Downey, Donald J. Meyer, Rumi Kato Price, Edward L. Spitznagel

Research output: Contribution to conferencePaper

29 Scopus citations

Abstract

As artificial neural networks continue to find usefulness in fields which historically favor more traditional statistical methods, the neural practitioner inevitably learns of useful techniques well known to statisticians which have yet to find widespread use in the field of neural networks. One such method, commonly used in medical screening and diagnosis, is receiver operating characteristic (ROC) analysis. ROC analysis is easily applied to a neural classifier, yet today is rarely used to assess the performance of neural classifiers outside of the medical and signal detection fields. In this paper we show how ROC analysis can be applied to neural network classifiers and demonstrate its usefulness by applying it to the diagnosis of psychiatric illness. Benefits of ROC analysis include a more robust description of the network's predictive ability and a convenient way to `tune' an already trained network according to differential costs of misclassification and varying prior probabilities of class occurrences.

Original languageEnglish
Pages3642-3646
Number of pages5
StatePublished - Dec 1 1999
Externally publishedYes
EventInternational Joint Conference on Neural Networks (IJCNN'99) - Washington, DC, USA
Duration: Jul 10 1999Jul 16 1999

Conference

ConferenceInternational Joint Conference on Neural Networks (IJCNN'99)
CityWashington, DC, USA
Period07/10/9907/16/99

Fingerprint Dive into the research topics of 'Using the receiver operating characteristic to assess the performance of neural classifiers'. Together they form a unique fingerprint.

  • Cite this

    Downey, T. J., Meyer, D. J., Price, R. K., & Spitznagel, E. L. (1999). Using the receiver operating characteristic to assess the performance of neural classifiers. 3642-3646. Paper presented at International Joint Conference on Neural Networks (IJCNN'99), Washington, DC, USA, .