What makes tourists feel negatively about tourism destinations? Application of hybrid text mining methodology to smart destination management

As an aspect of smart tourism, this study applied the sentiment analysis method to analyze travelers' online reviews of Paris. A total of 19,835 pieces of review data collected from a traveler review site (www.virtualtourist.com) were processed. All reviews were grouped into 14 categories as follows: overview, restaurants, sightseeing, hotels, things to do, night life, transportation, shopping, sporting & outdoors, favorites, off the beaten path, what to pack, tourist traps, warnings and danger, and local customs. Tourists' perception about the service in each category was successfully measured, and as an illustration, we chose “transportation” category that reported relatively low level of service quality for post-hoc analysis to reveal why tourists feel negatively about the transportation service.