Using big data to analyse tourism experiences

IPM visits Thailand

Following the tourist boat incidents that took place in July 2018, which claimed the lives of over forty Chinese tourists, the island of Phuket saw a dramatic decrease in visitor numbers. As one of the top tourist destinations in Thailand, and with an economy heavily dependent on tourism generated revenue, officials in Phuket now have the challenge of formulating plans to improve the experience of visitors in order to regain trust and increase loyalty.

Working in collaboration with Mahidol University International College (Bangkok), IPM visited Thailand in July to run two workshops that aimed to introduce tourism practitioners and academics to how machine learning using big data can help to turn around this decline. Funded through Manchester Metropolitan University’s Global Challenges Research Fund, the first session was held in Bangkok on Friday July 19th, with the second taking place in Phuket on Monday July 22nd.The workshops were facilitated by long-term IPM friend and collaborator Asst.Prof. Dr. Viriya Taecharungroj (Mahidol University International College). Dr Taecharungroj has been investigating how data from the review website TripAdvisor can be used to uncover which aspects of a place are in need of improvement, so that officials can apply effort and resource to the areas that are underperforming.

Over recent years, TripAdvisor has become the most popular review platform in the tourism industry for users and researchers alike. According to its official website, by the end of 2018 TripAdvisor had an average of more than 490 million monthly visitors and a total of more than 730 million reviews and opinions. The remarkable emergence of review platforms such as TripAdvisor has led to a widespread change in reputation management by organisations for which understanding tourists is the central issue. The purpose of Dr. Taecharungroj’s research is to develop a methodology that can analyse online reviews using machine learning techniques in such a way that practitioners in the fields of tourism and destination management can understand and apply these techniques to improve their attractions. The research studies the reviews of attractions including beaches, islands, temples, a pedestrian street, and markets in Phuket. In total, 65,079 online reviews were analysed using two machine learning techniques: latent Dirichlet allocation (LDA) and naı ̈ve Bayes modelling.

Attendees at the workshops were given an introduction into the methodology, including the statistical algorithms employed, with a live demonstration of how large data sets can be processed to produce meaningful outputs that can be utilised to the benefit of all manner of places. It is hoped that the methodologies proposed can now be utilised by destination marketing officials in order to develop practical suggestions for the sustainable improvement of the visitor experience in Thailand.  

Given IPM’s longstanding endorsement of evidence-based decision-making, we are excited to see how this work can be developed to help more places. As such, Dr Taecharungroj will be visiting IPM for a period of five months with a view to refining the methodologies and widening the applicability of the research techniques to towns and cities in the UK.

In addition to Dr Taecharungroj and colleagues at Mahidol University International College, who hosted us in Bangkok, IPM would like to thank Manchester Metropolitan University (Global Challenges Research Fund/UCRKE) for funding the workshop, and The Wyhdham Grand Phuket for hosting the Phuket workshop.

IPM Members can access more of Dr Taecharungroj’s work free of charge through the Journal of Place Management & Development (Volume 12, Issue 1).