Modelling of Noise-induced Annoyance: A Neuro-fuzzy Approach

An attempt has been made in this paper to develop a neuro-fuzzy model to investigate the effects of noise pollution on annoyance. A neuro-fuzzy model inherits the interpretability of fuzzy models and learning capability of neural networks in a single system. They have got wide acceptance for modelling many real world problems because it provides a systematic and directed approach for model building and gives the best possible design parameters in minimum time. The data used in the present model are synthetically generated from the fuzzy model developed by the author on the basis of survey findings of World Health Organization (WHO). This is implemented on Fuzzy Logic Toolbox of MATLAB using Adaptive Neuro-fuzzy Inference System (ANFIS). Out of the total input/output data sets, 80% was used for training and 20% for checking to validate the model. The annoyance has been considered as a function of noise level, its duration of occurrences, and socio-economic status of a person. The results of the model are found to be in good agreement with the findings of WHO on noise-induced annoyance.