A backpropagation neural network for risk assessment

Ray R. Hashemi, Nancy L. Stafford

Research output: Contribution to book or proceedingConference articlepeer-review

5 Scopus citations

Abstract

The authors investigate the credibility of the neural network approach as a viable tool in the field of developmental toxicity risk assessment. A three-layer artificial neural network (ANN) was trained using backpropagation. The topology of the network was decided based on a set of trials and errors. This network was trained to perform risk assessment on a set of toxicological data and give a decision like the decision given by experts. The assessment ability of the resulting network was compared with the statistical approach of discriminant analysis and the superiority of the neural network approach was established.

Original languageEnglish
Title of host publicationProceedings of Phoenix Conference on Computers and Communications, PCCC 1993
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages565-570
Number of pages6
ISBN (Electronic)0780309227, 9780780309227
DOIs
StatePublished - 1993
Event1993 Phoenix Conference on Computers and Communications, PCCC 1993 - Tempe, United States
Duration: Mar 23 1993Mar 26 1993

Publication series

NameProceedings of Phoenix Conference on Computers and Communications, PCCC 1993

Conference

Conference1993 Phoenix Conference on Computers and Communications, PCCC 1993
Country/TerritoryUnited States
CityTempe
Period03/23/9303/26/93

Scopus Subject Areas

  • Computational Theory and Mathematics
  • Computer Networks and Communications
  • Hardware and Architecture
  • Software

Fingerprint

Dive into the research topics of 'A backpropagation neural network for risk assessment'. Together they form a unique fingerprint.

Cite this