Data Mining from Big Data Analysis (D5.3)
Anastasia Arabadzhyan, Paolo Figin, Laura Vic · Zenodo (CERN European Organization for Nuclear Research) · 2021
This Report is the main outcome of Task 5.3, of Soclimpact project, which aimed at experimenting with different types of large datasets related to the tourism sector and characterised by velocity, variety, and volume, hence sharing the typical properties of Big Data. Three different lines of investigation were developed, aiming at tackling the following research questions: - How do climate events (like heat waves or storms, proxied by weather conditions and forecasts) impact on accommodation prices? This analysis considered Corsica, Sardinia and Sicily. - How do climate events impact on destination image and on the activities carried out by tourists in the destination? This analysis considered four islands of the Canary Islands archipelago (Gran Canaria, Fuerteventura, Lanzarote and Tenerife) and four Mediterranean islands (Crete, Cyprus, Malta and Sicily). - How do climate events (like forest fires) impact on hotel performance? This analysis considered Gran Canaria. Once estimations of impacts of specific events on prices, performance, and destination image were generated, climate change scenarios were used as inputs, thus allowing for the computation of the economic impact of climate change on the tourism sector (measured in terms of variation in yearly tourism expenditure). This report explains the methodologies used in the different lines of investigation, the estimation of parameters, and how to connect them with scenarios of climate change to produce estimations of changes in tourism expenditure in the different islands under scrutiny. Finally, the limitations of this work and the potentialities of Big Data analysis for research on the socio-economic impact of climate change are presented and critically discussed.