Loading...
Forecasting tourism demand through search queries and machine learning
de Kort, Rendell
de Kort, Rendell
Citations
Altmetric:
Keyword
Forecast combination
machine learning
feature selection
tourism demand forecasting
random forest
search data
machine learning
feature selection
tourism demand forecasting
random forest
search data
Location research
Aruba
Date
2017
Language
English
ISSN
ISBN
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
This paper utilizes different machine learning techniques for tourism demand forecasting. Considering the magnitude of tourism in terms of economic contribution to Small Island Developing States (SIDS), policy making could benefit greatly from accurate tourism demand forecasting. This paper pursues a novel approach of identifying relevant search query features through google correlate and applying machine learning techniques to estimate individual source market series prior to aggregation. The prediction performance of several machine learning methods is assessed when applied to monthly tourist arrivals from individual source countries to Aruba from 1994 to 2016. The results indicate that machine learning techniques in combination with novel internet datasets sets pose great potential for achieving accurate tourism demand forecasts.
Citation
De Kort, R. (2017), Forecasting tourism demand through search queries and machine learning. At IFC-Bank Indonesia Satellite Seminar on Big Data at ISI Regional Statistics Conference 2017, IFC-Bank Indonesia
Sponsorship
Publisher
Journal
Target group
URI
https://hdl.handle.net/20.500.14473/919