Diabetes mellitus forecast using Artificial Neural Network (ANN) / Siti Farhanah Jaafar

This project report presents the application of Artificial Neural Network (ANN) for forecasting the diabetes mellitus. The main objectives of this project are to forecast whether someone is the diabetes sufferer or not. The back-propagation algorithm of ANN has been chosen to train and test the data. Lots of studies carried out by many academia shows the performance of the neural network in predicting clinical outcomes accurately. Inputs of analysis are number of times pregnant, plasma glucose concentration, blood pressure, triceps skin fold thickness, serum insulin; body mass index, pedigree and age. The network with seven inputs is then tested and results obtained are compared in terms of analysis errors, number of inputs, number of layers and learning parameters.