In this project we study information needs of people that are interested in group tours as well as interactions between them. Compared to other travel related products, customers require more heterogeneous and detailed information when booking a group tour; the decision making process usually takes longer.
In our study we analyze user-generated content from the Web platform TourRadar.com. On this platform, group tours all over the world can be compared and booked. Furthermore, users are given the opportunity to engage with co-travelers in so-called meets before the tour starts. In those meets typically tour related questions are discussed. After the tour reviews can be published on the platform. A review may contain free text as well as five-star ratings for a number of categories such as value for money, guide and transportation. From both meets and reviews we extract topics that are relevant to the users with the help of text mining approaches, topic models and machine learning techniques. Furthermore, we apply social network analysis to study the user community of the platform and their dynamics.
The results of our analyses will be used to enhance the information content on the platform in order to facilitate the decision making process of the customers.