An application of artificial intelligence techniques in prediction of birds soundscape impact on tourists’ mental restoration in natural urban areas

Abstract

The characteristics of birds' sounds assume a primary role in tourists' mental restoration and stress recovery. The aim of this research is the evaluation of birds' sound composition in mental restoration of urban tourists to develop a decision support system as a practical tool. In this order, the recorded sounds of six birds were composed (57 composed sounds) and the human perception approach was used to assess the impact of sounds on the urban tourist's mental restoration. The MLP (Multi-Layer Perceptron), RBFNN (Radial Basis Function Neural Network) and SVM (Support Vector Machine) models were developed for mental restoration prediction in different birds' sound compositions. The results indicated that RBFNN model output (R2 training = 0.89, and R2 test = 0.85) has the best accuracy compared to the MLP and SVM models in prediction of birds' soundscape score in natural urban areas. According to the sensitivity analysis, the values of White eared Bulbul (Pycononotus leucotis), Great Tit (Parus major), House Sparrow (Passer domesticus), Laughing Dove (Spilopelia senegalensis), White Wagtail (Motacilla alba), and Eurasian Magpie (Pica pica) are prioritized respectively that influence the RBFNN model outputs. In practice, the designed environmental decision support system tool is applied by urban planners, managers, psychoacoustic researchers, and landscape architects to predict the landscape score in different birds' habitats.