Harnessing Machine Learning and Ensemble Models for Tourism Potential Zone Prediction for the Assam State of India

Abstract

Although Assam is enriched with several popular tourist destinations but till date, its’ complete charm remains enigmatic. This research was aimed at prognosticating the Tourism Potential Zone (TPZ) for the state of Assam using five machine learning (i.e., Conditional Inference Tree, Bagged CART, Random Forest, Random Forest with Conditional Inference Tree, and Gradient Boosting models) and one ensemble model. A 5-step methodology was implemented to do this research. First, a Tourism Inventory Database was prepared using the Google earth Imagery, and a rapid field investigation carried out with the help of Global Positioning System and non-participant observation technique. Total 365 tourism points was in the inventory, 70% (224) of which was used for the training set and 30% (124) was used for the validation purpose. The tourism conditioning factors such as Relief, Aspect, Viewshed, Forest Area, Wetland, Coefficient of Variation of rainfall, Reserve Forest, Population Density, Population Growth Rate, Literacy Rate and Road-railway density were used as the independent variables in the modelling process. The TPZ was predicted with the help of above machine learning models and finally, a new TPZ Ensemble Model was proposed by combining each model. The result showed that all machine learning models performed well according to prediction accuracy and finally, the ensemble model outperformed other models by achieving the highest AUC (97.6%), Kappa (0.82) and accuracy (0.93) values. The results obtained from this research using machine learning and ensemble methods can provide proper and significant information for decision makers for the development of tourism in the region.