Predictive Modeling of Tourist Satisfaction Based on Service Marketing Mix Elements Using Machine Learning Techniques

Area of Science:

• Tourism Marketing
• Service Marketing
• Behavioral Economics

Background:

• Tourist satisfaction and loyalty are key metrics for destination success.
• Understanding the influence of the service marketing mix is vital for tourism development.
• Cox's Bazar, a popular tourist destination, requires strategic marketing insights.

Purpose of the Study:

• To investigate the impact of the service marketing mix on tourist satisfaction and loyalty in Cox's Bazar.
• To identify the most influential marketing mix elements for enhancing tourist experience.
• To develop a robust predictive model for tourist satisfaction.

Main Methods:

• Data collection from 500 respondents using surveys.
• Statistical analysis including Cronbach's alpha, Variance Inflation Factor (VIF), and Principal Component Analysis (PCA).
• Machine learning models (XGBoost) for predictive analysis, validated through cross-validation, sensitivity, and learning curve evaluations.

Main Results:

• XGBoost model demonstrated high predictive accuracy (R-squared = 0.74, MSE = 0.10).
• The aggregated price and place (location) variable was the most significant predictor (68.20%).
• Cronbach's alpha values exceeded 0.70, confirming construct reliability; VIF was below 1.05, indicating no multicollinearity.

Conclusions:

• Pricing and location strategies are paramount for improving tourist satisfaction and loyalty in Cox's Bazar.
• The study validates the effectiveness of the XGBoost model for tourism marketing research.
• Findings provide actionable recommendations for policymakers and tourism stakeholders to optimize marketing efforts and enhance the tourist journey.