Topic modelling for wildlife tourism online reviews: analysis of quality factors

ABSTRACT

Improving the wildlife tourism experience of travellers is critical to achieving environmental sustainability. Therefore, it’s critical to investigate quality factors to raise experience quality. This study aims to identify the most salient quality factors and their relative importance to the wildlife tourism experience of travellers. In order to take advantage of big data and unprompted textual data, a machine learning method, namely, supervised latent Dirichlet allocation, was used in analysing online reviews of wildlife tourism. A total of 18 quality factors are extracted from data and cross-validated by comparing a prior theoretical framework. Results reveal that travellers’ perception of intensity, authenticity, species popularity, and operators’ rule management are crucial for enhancing traveller’s experience. In addition, the negative influences of wildlife attack and disease highlight the importance of control of encounter. These findings suggest some future research directions and provide practical implications.

KEYWORDS:

• Quality factors
• topic modelling
• supervised LDA
• wildlife tourism experience
• importance analysis

Disclosure statement

No potential conflict of interest was reported by the author(s).