PENERAPAN NEURAL NETWORK BERBASIS BACKWARD ELIMINATION UNTUK PREDIKSI SUSU SAPI PERAH

Abstract

Predicted milk production of dairy cows is one way that can be used to determine the condition of food supplies. Although there has been research on the condition of food, research begins with a data processing milk production of dairy cows to get a time series of data of milk production of dairy cows. The method used is to compare methods of neural networks with methods of artificial neural network-based backward elimination with the input data is the data on the previous year. Several experiments were conducted to obtain the optimal architecture and generate accurate predictions.

The results show the method of artificial neural network-based backward elimination produces smaller RMSE 0272 compared to only using the neural network only.