Using the ADAP Learning Algorithm to Forecast the Onset of Diabetes Mellitus

摘要: Abstract Neural networks or connectionist models for parallel processing are not new. However, a resurgence of interest in the past half decade has occurred. In part, this is related to better understanding what now referred as hidden nodes. These algorithms considered be marked value pattern recognition problems. Because that, we tested ability an early neural network model, ADAP, forecast onset diabetes mellitus high risk population Pima Indians. The algorithm's performance was analyzed using standard measures clinical tests: sensitivity, specificity, and receiver operating characteristic curve. crossover point sensitivity specificity 0.76. We currently further examining these methods by comparing ADAP results with those obtained from logistic regression linear perceptron precisely same training forecasting sets. A description algorithm included.