A Neural Network based Approach for Predicting Indonesian Teacher Engagement Index (ITEI)

The Indonesian Teacher Engagement Index (ITEI) is an instrument designed to help teachers detect themselves through self-diagnostics. In this study we use the Neural Network approach to predict Index values and assess parameters in the Neural Network. The analysis of this study is based on several variables including the input layer of the proposed ANN model as many as 28 parameters obtained from the ITEI application related to the teacher index value. After training our network we obtain a Root Mean Square Error of around 0.2%. The best model that can be suggested by the Neural Network to sample the Indonesian Teacher Engagement Index (ITEI) is in the condition when the Learning rate value: 0.3, momentum: 0.5, Number of Hidden Layer: 2 where the number of Neurons in layer 1 is 6 and Neuron in layer 2 is equal to 6.