Table 1 A comparison of the related studies based on tourism forecasting.
From: Predicting tourism growth in Saudi Arabia with machine learning models for vision 2030 perspective
Study | Region | Used dataset | Method(s) | Data period | Key findings | Limitations |
|---|---|---|---|---|---|---|
Global | A dataset for most of countries between 2010 to 2020 | SVR, RF | 2010–2020 | SVR achieved 99.4% accuracy | Lacks regional/contextual focus | |
Global | Private Dataset | SVM, RF, DT | Annual & Quarterly | SVM performed best on quarterly data | No integration of external drivers | |
Global | Private Dataset | ANN, RNN, LSTM | Multi-year | LSTM superior for motivational features | Limited interpretability | |
Global | Collected data from Google Trends | AFSO-DGRU | Web search data | MAPE 1.34 | Requires high-quality search data | |
Saudi Arabia | HAJJv2 dataset-based videos | Gradient Boosting | Hajj Data | 87% accuracy, 93% precision, 87% recall, and 84% F1-score; | Not a forecasting model | |
Saudi Arabia | Private Datasets from the Saudi Tourism Authority (2015–2021) | RF, DT, SVM, GNB | 2015–2021 | Gaussian Naive Bayes reached 99.993%; Accurate spending prediction | No demand forecasting | |
Saudi Arabia | Data obtained from the Saudi Tourism Information and Research Centre (MAS) (2000–2019 | Gravity Models | 2000–2019 | Positive economic impact | Limited interpretability | |
Saudi Arabia | Collected data from the General Authority for Open Data statistics | Liner regression, Naive Baye | 2020–2023 | Naive Baye had the least deviation; | No demand forecasting |
