User satisfaction in rural ecotourism accommodations has become a critical determinant of success in the tourism industry. With the rapid growth of online booking platforms such as Jabama and intensifying competition, data-driven analysis is increasingly essential for identifying the factors influencing satisfaction. Despite the growing popularity of ecotourism in Iran, few studies have applied advanced and interpretable machine learning methods to explore this topic. The rise of the sharing economy and digital accommodation platforms has transformed user experiences, emphasizing the importance of evidence-based insights.

This study collected data from 1,123 rural ecotourism lodgings listed on the Jabama platform using Python and the Selenium library. Independent variables included information quality, cleanliness, value for money, check-in experience, hosting quality, and location, while user satisfaction ratings served as the dependent variable. Regression-based models—linear regression, decision tree, random forest, gradient boosting, and support vector regression (SVR) with an RBF kernel—were implemented. Model performance was evaluated using mean squared error (MSE), the coefficient of determination (R²), and 5-fold cross-validation to ensure reliability and robustness.

The results showed that random forest and gradient boosting achieved the highest accuracy, with R² values above 0.86 and MSE below 0.012. Feature importance analysis revealed that information quality (importance score = 0.442), cleanliness, and value for money were the strongest predictors of user satisfaction. The decision tree model provided interpretable decision rules, highlighting the central role of information quality at the root level and the subsequent influence of cleanliness and value for money. These findings underscore the potential of interpretable machine learning approaches for enhancing user satisfaction analysis in Iran’s ecotourism sector and guiding data-informed managerial decisions.