Will you like this place? A tag-based place representation approach

Abstract

Increasingly popular location-based services can monitor our geographical positions in real time and so can provide a fundamental source for capturing users feelings and personal attitudes towards a particular place at a particular time. We propose a novel procedure for the representation of places through weighted tag-lists based on user reviews on these type of services. In our method the resulting lists can be built according to different criteria aiming to highlight differences and similarities among locations that can be in geographical proximity, belong to a similar type/category, or be included in the personal mobility history of a specific user.