Analisis dan Perancangan Jaringan Saraf Tiruan untuk Mengidentifikasi Tingkat Kematangan Buah Belimbing Manis (Averrhoa carambola L.)

Abstrak

Starfruit is one of the fruits that are widely cultivated in Indonesia. But at this time sorting of starfruit is still
done manually by humans, consequently resulting in a uniform level of maturity that is not good. For this
reason, a system is needed that can identify the level of maturity of starfruit with artificial neural networks.
The main problem of designing artificial neural networks is how to analyze and design an artificial neural
network architecture in order to determine the maturity level of sweet starfruit properly. This study aims to
design artificial neural networks with backpropagation method to identify the maturity level of starfruit.
From the results of the study, the best configuring of backpropagation artificial neural network model is a
model of artificial neural networks with 3 inputs, 11 hidden layer neurons and 3 outputs (3-11-3). With this
configuration, artificial neural networks are able to identify the level of maturity with a success rate of 95.8%
of 48 starfruit test data.