Discovering daily POI exploitation using LTE cell tower access traces

Abstract

Point of interest (POI) in a urban space represents the perception of city dwellers and visitors with respect to a certain place. Among many physical data sources, LTE mobile traffic patterns are used to derive POIs. In this paper, we propose a POI discovery scheme that can extract not only diachronic POIs in a given place but also how they are exploited everyday from LTE cell tower access data. We extract meaningful features from LTE cell tower access data and build a machine learning based model which is fitted to our domain. Our model shows better performance compared to existing works. We show that LTE data is a very powerful data source for interpreting places, and this model will be very helpful for building urban space-based services in the future.