Mining Tourism Experience on Twitter: A case study (2207.00816v1)

Abstract: With the increase of digital data and social network platforms the impact of social media science in driving company decision related to product/service features and customer care operations is becoming more crucial. In particular, platform such as Twitter where people can share experience about almost everything can drastically impact the reputation and offering of a company as well as of a place or tourism site. Text mining tools are researched and proposed in literature in order to gain value and perform trend topics and sentiment analysis on Twitter. As data are the fuels for these models, the "right" ones, i.e the domain-related ones makes a difference on their accuracy. In this paper, we describe a pipeline of \textit{DataOps / MLOps} operations performed over a tourism related Twitter dataset in order to comprehend tourism motivation and interest. The gained knowledge can be exploit, by the travel/hospitality industry in order to develop data-driven strategic service, and by travelers which can consume relevant information about tourist destination.