Evaluation of Tourist Satisfaction Based Light Gradient Boosting Machine Technique

In recent years, tourist satisfaction has become a focus as the population has become a key factor in the economic development of those regions. A good understanding of what exactly needs to be done to improve public transport services is key to a city’s attractiveness. The research aimed at a sample population of 1019 tourists, consisting of foreign origin and local residents, through a stratified sampling method and open-ended survey questions. Discussing data mining, for example, multidimensional analysis and clustering visualization, intersects with survey data to generate insights. Furthermore, a supervised light gradient boosting (LGBM) algorithm of Machine Learning (ML) is proposed to develop a predictive model that assigns tourists to satisfied, neutral, and dissatisfied categories, respectively. A highly successful model will be developed as a learning-facilitating framework and serve as a predictive tool for audience satisfaction ratings. The proposed method obtains the more efficient in accuracy of 97.4%, precision of 97.5%, recall of 97.4% and specificity of 98.9% from the existing methods.