Measuring and Recommending Time-Sensitive Routes from Location-Based Data

Abstract

Location-based services allow users to perform geospatial recording actions, which facilitates the mining of the moving activities of human beings. This article proposes to recommend time-sensitive trip routes consisting of a sequence of locations with associated timestamps based on knowledge extracted from largescale timestamped location sequence data (e.g., check-ins and GPS traces). We argue that 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 route goodness function that aims to measure the quality of a route. Equipped with the route goodness, we recommend time-sensitive routes for two scenarios. The first is about constructing the route based on the user-specified source location with the starting time. The second is about composing the route between the specified source location and the destination location given a starting time. To handle these queries, we propose a search method, Guidance Search, which consists of a novel heuristic satisfaction function that guides the search toward the destination location and a backward checking mechanism to boost the effectiveness of the constructed route. Experiments on the Gowalla check-in datasets demonstrate the effectiveness of our model on detecting real routes and performing cloze test of routes, comparing with other baseline methods. We also develop a system TripRouter as a real-time demo platform.

Geo-Social Media Analytics

“…(3) Location Recommendation [9,10,13,22,23]. Location recommendation is to recommend new locations that users have never visited before.…”

Point of interests recommendation in location-based social networks

“…Existing research on personalized POI recommendation mainly explores the geographic influence to improve the recommendation accuracy, based on the observation that the geographic proximity between spatial items affect users check-in locations [42,43]. Recently, there are works that further integrate social influence in LBSNs to recommend items as common interests shared by social friends [29]. In terms of the temporal effect of user check-in activities in LBSNs, most existing work only investigate the temporal cyclic patterns of checkins [22].…”

“…tend to go to restaurants at dinner time and then relax in cinemas or bars at night [87]), geographical proximity (e.g., tourists often sequentially visit London Eye, Big Ben and Downing Street [83]) and the coherence between human preference and the type of places (e.g., people usually check in at a stadium before a restaurant instead of the reverse way because it is not healthy to exercise right after a meal [29]). …”

“…While it has been shown in multiple studies that human movement in LBSNs clearly demonstrates sequential patterns [87,83,29], leveraging sequential patterns for spatial item recommendation has not been well-studied.…”

Mining user behavior in location-based social networks

“…Using similarity measures based on location histories is possible perform different mining tasks on LBSNs, such as classification MUSOLESI, 2014;YU et al, 2015), community detection (BROWN et al, 2012;WANG et al, 2014;, link prediction (MENG-SHOEL et al, 2013;BAYRAK;POLAT, 2014;LIN, 2015;KYLASA;KOLLIAS;GRAMA, 2016), among others. As discussed in previous chapters, link prediction is commonly used due to its ability to capture both user and relationship patterns and so identify the different ways in which network actors could to establish new relationships (LÜ; KONG;YU, 2014;.…”