Modelling and Analysis of Salmonella Typhimurium Infections using Logistic Regression and Neural Network Models

Analysis of the risk factors is very important to develop appropriate prevention and control strategies for Salmonella Typhimurium infections. In this paper, basic case-control analysis, logistic regression models and neural network models are developed to identify the risk factors. The odds ratios and p values obtained by the neural network model are more credible in comparison to the case-control study and logistic regression model. The performance between logistic regression and neural network models are compared in terms of the mean absolute error, standard deviation of mean absolute error, correlation coefficient, and classification rate. The continue datasets (e.g., age, education) could be introduced into this model except binomial data in future study.