Abstract
Recommender systems have radically changed the way people find products, services and information. They are a precious tool in e-commerce and other online services and have slowly been clawing their way into the real-world stage. Location is one of the variables that can be useful in this new situation. While this particular area has been the subject of some research, it can go even further with the exploration of mobility. In this work, we analyze the integration of mobility in a recommender system with real mobility data from a public transportation network. We developed an algorithm that incorporates location and frequency in a conventional recommender system. Our results show successful recommendations of items adapted to users' mobility patterns.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have