Research on factors affecting tourist involvement in coffee tourism after the COVID-19 pandemic in Thailand

Abstract

Background: The world economy was broken by the COVID-19 pandemic, which affected the coffee industry. The COVID-19 pandemic's financial effects might influence equity markets and personal lives. This includes financial commodities like coffee, which the pandemic is predicted to damage. Coffee tourism is an emerging new kind of tourism in Thailand, formed in response to growing demand from visitors with a particular affinity for the beverage. Coffee tourism may contribute considerably to the expansion of Thai tourism if given the proper guidance and assistance. Methods: As part of a coffee tourism experience focusing on first-hand activities and information, tourists can visit neighbouring sites while on a coffee plantation. This research uses a stochastic neuro-fuzzy decision tree (SNF-DT) to analyse coffee tourism in Thailand. The research surveys 400 international and Thai coffee tourists. According to studies, Thai visitors mostly visit coffee tourism locations in Thailand for enjoyment. They also wanted to visit coffee fields in order to get personal knowledge of coffee production and marketing. Based on the comments of Thai visitors, coffee tourism in northern Thailand looks to be highly and effectively handled. Due to the same factor, responses from foreign coffee tourists indicated that many of their journeys to coffee tourism destinations were made entirely for enjoyment rather than the business. They also wanted to meet local tour guides and acquire handmade and locally produced things to understand more about coffee tourism.  Result: According to study results, coffee tourism management in northern Thailand looks well-received by international tourists. We also compare the suggested model to the traditional one to demonstrate its efficacy. The performance metrics are prediction rate, prediction error, and accuracy. The estimated results for our proposed technique are prediction rate (95%), prediction error (97%), and accuracy (94%).