Forecasting daily attraction demand using big data from search engines and social media

Abstract

Purpose This paper aims to compare the forecasting performance of different models with and without big data predictors from search engines and social media. Design/methodology/approach Using daily tourist arrival data to Mount Longhu, China in 2018 and 2019, the authors estimated ARMA, ARMAX, Markov-switching auto-regression (MSAR), lasso model, elastic net model and post-lasso and post-elastic net models to conduct one- to seven-days-ahead forecasting. Search engine data and social media data from WeChat, Douyin and Weibo were incorporated to improve forecasting accuracy. Findings Results show that search engine data can substantially reduce forecasting error, whereas social media data has very limited value. Compared to the ARMAX/MSAR model without big data predictors, the corresponding post-lasso model reduced forecasting error by 39.29% based on mean square percentage error, 33.95% based on root mean square percentage error, 46.96% based on root mean squared error and 45.67% based on mean absolute scaled error. Practical implications Results highlight the importance of incorporating big data predictors into daily demand forecasting for tourism attractions. Originality/value This study represents a pioneering attempt to apply the regularized regression (e.g. lasso model and elastic net) in tourism forecasting and to explore various daily big data indicators across platforms as predictors.

Daily hotel demand forecasting with spatiotemporal features

“…It is popularly used in situations where obtaining acceptable results is possible owing to the simplicity of the basic structure of the data. Econometric models are used to identify and quantify the influence of explanatory variables on demand Tian et al (2021). Autoregressive distributed lag mode (Song et al , 2011), vector autoregression (Assaf and Tsionas, 2019) and mixed-data sampling (Zhang et al , 2021) have achieved superior performance in the existing studies.…”

Artificial intelligence and big data: ontological and communicative perspectives in multi-sectoral scenarios of modern businesses

“…The capacity of human beings to reason and rationally argue has remained unchanged in millions of years. The combined intellect of neuroscientists, engineers and mathematicians has led the modern AI and big data applications toward solving every issue in various sectors of the economy like education (Chaurasia et al, 2018), health (Anam and Haque, 2020), tourism (Tian et al, 2021), hospitality (Zarezadeh et al, 2022), supply chain management, (Del Giudice et al, 2021) etc. In the area of financial services also AI is being used extensively and the onus of development as well as deployment of AI for marketing such financial services lie on managers majorly (Mogaji and Nguyen, 2021).To adopt AI systems in industry government interventions at various levels are required to improve their implications for the society as a whole (Chatterjee, 2020;Samsurijan et al, 2022).…”

“…Gartner defines big data as “high-volume and/or high-variety information assets that demand cost-effective, innovative forms of information processing that enable enhanced insight, decision-making and process automation.” To put it simply, big data includes every means and practice/application that help people use and manage huge data sets. The concept of big data is heavily used in capturing trends, preferences and user behavior when people interact with various systems as well as one another (Zarezadeh et al , 2022; Del Vecchio et al , 2020; Mariani et al , 2021; Almeida, 2022; Tian et al , 2021; Simović, 2018; Sedkaoui and Khelfaoui, 2019). Big data can help companies analyze and figure out the motivations of their most important clients, while also providing ideas for the creation of new offerings (Mariani et al , 2018).…”

Exploring governance effectiveness, tourism development and poverty reduction relationship in SAARC countries using panel dynamic estimation

“…The Pedroni test permits slope coefficients to differ between countries and presented seven different types of tests to ensure slope heterogeneity across the countries. The Kao Residual cointegration test assumes panel homogeneity and cross-sectional independence (Tian et al, 2021). Pedroni and Kao's test is based on the null hypothesis of no cointegration and the alternative hypothesis of panel cointegration.…”