Stacking Ensemble Learning-Based Convolutional Gated Recurrent Neural Network for Diabetes Miletus

Diabetes mellitus is a metabolic disease in which blood glucose levels rise as result of pancreatic insulin production failure. It causes hyperglycemia and chronic multiorgan dysfunction, including blindness, renal failure, cardiovascular disease, if left untreated. One the essential checks that are needed to be performed frequently Type 1 Mellitus test, this procedure involves extracting quite frequently, leads subject discomfort increasing possibility infection when often recurring. Existing methods used for diabetes classification have less accuracy suffer from vanishing gradient problems, overcome these issues, we proposed stacking ensemble learning-based convolutional gated recurrent neural network (CGRNN) Metamodel algorithm. Our method initially performs outlier detection remove data, using Gaussian distribution method, Box-cox correctly order dataset. After outliers’ detection, missing values replaced by data’s mean rather than their elimination. In base model, multiple machine learning algorithms like Naïve Bayes, Bagging with random forest, Adaboost Decision tree been employed. CGRNN Meta model uses two hidden layers Long-Short-Time Memory (LSTM) Gated Recurrent Unit (GRU) calculate weight matrix prediction. Finally, calculated passed softmax function output layer produce prediction results. By LSTM-based CG-RNN, square error (MSE) value 0.016 obtained 91.33%.