Comparison Neural Network and Support Vector Machine for Production Quantity Prediction

This research presents the comparison of Neural Network (NN) model and Support Vector Machine (SVM) for predicting production quantity. The experiment is performed by optimizing the parameter of NN model and SVM by trial and error to find the smallest error between actual and predicted. The private data set (foundry manufacturing) has been collected from 2006 until 2013 that used for testing such both models. This data set is split into training data and testing data using a cross-validation method for training and testing phase. The result shows that the NN model is better than SVM by having the smallest RMSE of the NN model was 1.713, meanwhile SVM was 1.718 on the training phase. After used in testing phase, can obtain RMSE of the NN model was 0.206284249 and for SVM 0.309426374. Finally, this comparison provides contribution to knowledge where NN makes impressive performance then SVM for production quantity prediction