Tourism forecasting by search engine data with noise-processing

Abstract

In many studies, search engine data were efficient to analyze and forecast as an explanatory variable, including the tourism volumes predictions. However, the search data and the tourism volumes were always interfered by the noise. Without noise-processing, the predictive ability of search engine data might be weak, even invalid. As a method of noise-processing, Hilbert-Huang Transform (HHT) could deal with non-linear and non-stationary data. This study proposed a model with denoising and forecasting by search engine data, namely CLSI-HHT. The search queries were composited into an index first, then the noise were extracted from the index and tourism volumes sequences by HHT. The study further forecast the tourism volumes with the effective series. The results demonstrated that CLSI-HHT model outperformed the baselines significantly while the index model without denoising performs nearly same as the time series model. Moreover, wavelet transform and filtering were compared with HHT on denoising and the results implied that HHT had higher signal noise ratio (SNR) and forecast more accurately. The study concluded that noise-processing was necessary for the tourism forecasting with search engine data, and HHT could be an effective method on denoising.

Heat stroke internet searches can be a new heatwave health warning surveillance indicator

“…Most of the temperature and health outcomes are studied using mortality data while few use morbidity data, but the former is not well suited for representing general population health and it still leaves a high level of uncertainty in health risk prediction during heatwaves. Nowadays, the increasing availability of datasets from sources such as social media posts, search engine queries and other internet data have shown the potential for analyzing patterns, trends and social phenomena in a variety of domains including finance1617, science18, tourism1920 and health21222324. In this study, we conducted a correlation and linear regression analysis to test the relationship between heat stroke internet searches and heat stroke health outcomes in Shanghai, China, during the summer of 2013.…”

Google Trends data for analysing tourists’ online search behaviour and improving demand forecasting: the case of Åre, Sweden

“…While Pan et al (2012) utilize search engine data to enhance the forecast accuracy of hotel (i.e. room) demand, other studies are primarily focussing on the prediction of tourist arrivals (Li et al 2016;Höpken et al 2017). Finally, the study by Yang et al (2014) confirms the value of web traffic data from local destination marketing organizations (DMOs) in predicting the demand for hotel rooms in a tourist destination.…”

“…While Liu et al (2012) use mixed metrics for measuring the similarity between lagged predictors and the target time series variable (i.e. Pearson correlation and Kullback-Leibler divergences), other studies exclusively rely on Pearson correlation to identify significant lags (Yang et al 2015;Li et al 2016;Pan et al 2017). However, the reliability of the Pearson correlation coefficient is limited as it depends on statistical assumptions.…”

“…As search query data can also contain useless information, Li et al (2016) proposed a method for noise reduction of search query series which builds on the methodology by Yang et al (2015). The authors conclude that noise processing is an essential step in forecasting with Google Trends data.…”

“…The authors conclude that noise processing is an essential step in forecasting with Google Trends data. Typically, 'Hilbert-Huang-Transformation' (HHT) is applied for noise processing, which reduces prediction errors significantly (Li et al 2016). Peng et al (2017) recently made another progress.…”