The application of artificial neural network in studying landless farmer's mental health problems / 中华流行病学杂志

ABSTRACT

To introduce a method of classification with high precision-the artificial neural network (ANN),and to compare the results using logistic method. Using data from 1070 landless peasants'mental health survey,the artificial neural network models and logistic regression model were built and compared on their advantages and disadvantages of the two models.The prediction accuracy for artificial neural network was 94.229% and for logistic regression it was 51.028%. ANN appeared to have had good ability on generalization. ANN displayed advantages when conditions of classical statistical techniques could not be met or the predictive effect appeared to be unsatisfactory. Hence, ANN would make a better facture of its application in medical researche.