Exploiting large-scale check-in data to recommend time-sensitive routes

Abstract

Location-based services allow users to perform geo-spatial check-in actions, which facilitates the mining of the moving activities of human beings. This paper proposes to recommend time-sensitive trip routes, consisting of a sequence of locations with associated time stamps, based on knowledge extracted from large-scale check-in data. Given a query location with the starting time, our goal is to recommend a time-sensitive route. We argue a good route should consider (a) the popularity of places, (b) the visiting order of places, (c) the proper visiting time of each place, and (d) the proper transit time from one place to another. By devising a statistical model, we integrate these four factors into a goodness function which aims to measure the quality of a route. Equipped with the goodness measure, we propose a greedy method to construct the time-sensitive route for the query. Experiments on Gowalla datasets demonstrate the effectiveness of our model on detecting real routes and cloze test of routes, comparing with other baseline methods. We also develop a system TripRouter as a real-time demo platform.