A prediction study on tourist amount based on web search data

The web search data, which records hundreds of millions of searchers' concerns and interests, reflects the trends of their behavior and provides essential data basis for the prediction of tourist amount. In this paper, firstly, we build a systematic theoretical framework reveal the correlation between web search and tourists' travel. Then, at the basis of the theoretical framework, we establish a search index for forecasting the amount of visitors, and the empirical study on XIN JIANG-heavenly lake testes and verify the co-integration relationship between search index and tourist amount. Finally, we establish a prediction model based on both web search index and historical data. The results demonstrate that the Mean Absolute Percent Error(MAPE) of this model decrease from 4.46% to 1.81% comparing with the traditional auto-regression AR model when they are used to forecast the number of visitors for three weeks. The conclusions of this paper can be used as references for tourism-related authorities when they try to monitor the change of tourist amount and offer adequate tourism services. Moreover, this new prediction method considering search index can be applied to other web-based soc-economical activities.