Penerapan Jaringan Saraf Tiruan Propagasi Balik Studi Kasus Pengenalan Jenis Kopi

Bernard Renaldy Suteja

Abstract


Testing the taste of coffee traditionally, will depend on human taste sensor, which is not stable and relied on physical and mental conditions. In this situation, only qualitative measures can be determined. To increase reliability in the evaluation of taste of coffee, an electronic taste sensor system can be used to measure objectively. Neural network approach is very appropriate to apply pattern recognition method in an electronic taste sensor system. This paper discusses about neural network with back propagation for recognizing the kind of coffee. Back propagation architecture was chosen because can produce many tastes as output, so alternative identifications can be seen ( in the form of correlation factor).


References



Full Text: PDF

Refbacks

  • There are currently no refbacks.