Table 2.
Summary of selected tourism demand forecasting studies using search query data.
| Author (Year) | Data frequency | Data categories | Search data preprocessing | Analysis method | Predicted context |
|---|---|---|---|---|---|
| Li et al. (2018) | Monthly | Baidu Index | PCA | PCA-ADE-BPNN | Beijing in-bound tourist volume |
| Li and Law (2019) | Monthly | Google Trends | EEMD | ARX | Tourist arrivals from nine countries to Hong Kong |
| Liu, Zhang, Zhang, Sun, & Qiu (2019) | Daily | Baidu Index | NA | VAR | Tourist arrivals to Tianmu lake in China |
| Law et al. (2019) | Monthly | Google Trends | MIC | DL | Tourist arrivals in Macau |
| Hu and Song (2019) | Monthly | Google Trends | AIC | ANN | Short-haul travel from Hong Kong to Macau |
| Sun et al. (2019) | Monthly | Google Trends and Baidu Index | Correlation analysis | KELM | Tourist arrivals to Beijing |
| Wen and Liu (2019) | Monthly | Baidu Index | PCA | Hybrid models | Tourist arrivals in Hong Kong from mainland China |
| Li, Hu, and Li (2020) | Weekly | Baidu Index | GDFM | ARIMAX, SVM, and RF | Tourist arrivals to Mount Siguniang |
| Li, Li, et al. (2020) | Weekly and Monthly | Google Trends and Baidu Index | Machine learning-based feature selection models | ARMAX | Monthly domestic tourist arrivals to Beijing, China and weekly forecasting of hotel occupancy in Charleston, SC |
| Wen et al. (2020) | Daily | Baidu Index | GDFM | SARIMA-MIDAS | Tourist arrivals in Hong Kong from mainland China |
| Bi et al. (2020) | Daily | Baidu Index | Pearson correlation coefficient | LSTM | Daily tourism volume of Jiuzhaigou and Huangshan Mountain Area |
| Tang, Zhang, Li, and Li (2021) | Monthly | Baidu Index | Pearson correlation coefficient and PCA | SED-BEMD-LR, SED-BEMD-SARIMA, SED-BEMD-SVR, SED-BEMD-ELM, and SED-BEMD-RVFL | Tourist arrivals to Hainan, China |
| Xie, Li, Qian and Wang (2020) | Baidu Index and Google Trends | KPCA | ARIMAX, BPNN, SVR, LSSVR, MA-LSSVR | Tourist arrivals to Hong Kong from Chinese Mainland and United Stated | |
| Bi, Li, Xu and Li (2021) | Daily | Baidu Index | Correlation analysis | Ensemble LSTM with CPS | Daily tourism demand forecasting for the Huangshan Mountain Area |
| Hu, Xiao and Li (2021) | Weekly | Baidu Index | Boruta algorithm | ARIMAX | Weekly tourist arrivals to Mount Siguniang and Kulangsu |
| Xie, Qian, and Wang (2021) | Monthly | Baidu Index | Correlation analysis | LSSVR-GSA | The volume of cruise tourism in China |
| Tian, Yang, Mao and Tang (2021) | Daily | Baidu Index | Lasso and elastic net | ARMAX and MSAR | Daily tourist arrivals to Mount Longhu in China |
| Yang, Guo and Sun (2021) | Monthly | Baidu Index | Correlation analysis | LR | Domestic tourist arrivals to Chongqing |
| Sun et al. (2021) | Monthly | Google Trends | Pearson correlation coefficient | B-SAKE | Inbound tourist arrivals in Beijing from origin countries of United States, the United Kingdom, Germany, France |
| Yang et al. (2022) | Daily | Google Trends | LASSO | ARX | Tourism demand across 74 countries |
Notes: Adaptive differential evolution algorithm (ADE); Principal component analysis (PCA); Kernel principal component analysis (KPCA); Autoregressive exogenous model (ARX); Ensemble empirical mode decomposition(EEMD); Maximal information coefficient (MIC); Deep learning (DL); Akaike Information Criterion (AIC); Generalised dynamic factor model (GDFM); Bivariate empirical mode decomposition (BEMD); Least absolute shrinkage and selection operator (LASSO); Markov-switching auto-regression (MSAR); Random forest (RF); Least squares support vector regression model with gravitational search algorithm (LSSVR-GSA); Moving average–least squares support vector regression (MA-LSSVR); Bivariate empirical mode decomposition (BEMD); Seasonal autoregressive integrated moving Average–Mixed data sampling (SARIMA-MIDAS); correlation-based predictor selection(CPS).