An adaptive neuro-fuzzy approach to risk factor analysis of Salmonella Typhimurium infection

A clear understanding of risk factors is important in order to develop appropriate prevention and control strategies for infections caused by such pathogens as Salmonella Typhimurium. There have been many studies on modelling of pathogen infections by analysing the relevant risk factors. In this paper, a novel neuro-fuzzy based approach to analysing risk factors of Salmonella Typhimurium infections is proposed. The proposed approach incorporates neuro-adaptive learning techniques into the fuzzy logic method. Rather than choosing the parameters associated with a given membership function by trial and error, these parameters could be tuned automatically in a systematic manner so as to adjust the membership functions of the input/output variables for optimal system performance. A multi-factor predictive model is developed with 80% training data and the proposed approach is tested with the rest 20% unexposed data. The results demonstrate the effectiveness of the proposed approach in comparison to a typical fuzzy logic model.