This thesis explores various aspects of the destination recommendation domain, namely mobility analyses about which types of travelers can be observed in location-based social media data, how these groups perform in point-of-interest recommendation, how long travelers should stay at a destination, and which data sources are suitable to characterize destinations. Finally, we propose a city recommender system that supports users making the trade-offs involved in their travel choices.
The involvement of geographic information differentiates point-of-interest recommendation from traditional product recommendation. This geographic influence is usually manifested in the effect of users tending toward visiting nearby locations, but further mobility patterns can be used to model different groups of users.
To evaluate which data sources are most suitable to characterize destinations, we investigate 18 characterization methods that fall into the following categories: venue data, textual data, and factual data.
Tourism is a complex domain for recommender systems because of the high cost of recommending an unsuitable item and the absence of ratings to learn user preferences.
This is the first major publication of my dissertation project, which extends and combines previous work on the analysis of traveler mobility.