Using MLP Networks to Classify Red Wines and Water Readings of an Electronic Tongue

  • Authors:
  • Humberto Costa de Sousa;Antonio Riul Jr.

  • Affiliations:
  • -;-

  • Venue:
  • SBRN '02 Proceedings of the VII Brazilian Symposium on Neural Networks (SBRN'02)
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

Feasible efforts have been made to mimic the human gustatorysystem through an "artificial tongue". This devicecomprises an array of sensing units that is able to differentiatetastes with a higher sensitivity than the biological system.Experimental results indicate that when the data generatedby such sensing units are handled by artificial neuralnetworks, this "artificial tongue" can successfully discriminatewines of different winemakers, vintage and grapes, aswell as different brands of mineral water, distilled water andMilli-Q water. The accuracy achieved by the experimentssuggests that the sensing units may be used to detect abnormalchemical substances in a production line or even set anew approach to control quality standards in food industry.