Tripadvisor Users’ Profile Analysis: A Data Mining Approach

Nowadays, the tourism industry is one of the cornerstones of any sustainable economy, in a manner that it assigns about 10 percent of the global economy. The purpose of this article is to study and analyze the profile information of users of one of the most popular social tourism websites called Tripdvisor, by utilizing data mining techniques. For this purpose, the database of the tripadvisor website was considered and the profile information of the users who at least had commented on one of the Iranian hotels was extracted. After that all the information of user profiles were evaluated by feature selection technique and the information that had the most impact on the clustering were identified. In the following, by calculating the Davies–Bouldin index, the optimal number of clusters was obtained and it was three, and the users were divided into three clusters. Each cluster had unique characteristics that were named pioneers, aristocratic friends travel, and low-travel users. Then, based on the characteristics of each of these three clusters and their users' profile information by using association rules, each cluster was attempted to be further identified. Finally, proportional with the analyzed features, some solutions and approaches were provided to increase the participation of these users in the website and targeted promoting programs for each of these users' cluster.