<![CDATA[This research aims to explore the impact of various learning rate values in artificial neural networks (ANN) in increasing the accuracy of predicting rubber tree maintenance costs. Using a dataset that includes factors such as tree age, soil conditions, weather, and maintenance methods, an ANN model is built and tested with various learning rate values to find optimal parameters. The research results show that the influence of various learning rate values on the performance of artificial neural networks (ANN) in predicting rubber tree maintenance costs varies significantly. From the training and testing results, learning rate 0.1 shows the best results with MSE 0.00995387 and 75% accuracy on training data, and MSE 0.00976614 and 83% accuracy on testing data. This conclusion emphasizes the importance of choosing the right learning rate value in applying ANN to predict rubber tree maintenance costs, which is expected to help plantation managers improve operational efficiency and cost management.]]>
Copyrights © 2024
