A method for predicting future location of mobile user for location-based services system Thi Hong Nhan Vu a , Keun Ho Ryu a , Namkyu Park b, * a Department of Computer Science, Chungbuk National University, Cheongju, Republic of Korea b Department of Industrial Engineering, Ohio University, Stocker Center 277, Athens, OH 45701-2979, USA article info Article history: Available online 22 July 2008 Keywords: Trajectory Location prediction Movement rules Location-based services abstract Convergence of location-aware devices, wireless communications, and geographic information system (GIS) functionalities has been enabling the deployment of a new generation of selective information dis- seminating services and location-based services (LBSs). Current LBSs use information about current loca- tions of users to provide services, such as nearest features of interest, they request. Although the common computing strategy in LBSs benefits the users, there are additional benefits when future locations are pre- dicted. One major advantage of location prediction is that it provides LBSs with extended resources, mainly time, to improve system reliability which in turn increases the users’ confidence and the demand for LBSs. In this study, we propose a movement Rule-based Location Prediction method (RLP), to guess the user’s future location for LBSs. Its performance is assessed with respect to precision and recall. In com- parison with the previous technique, the prediction accuracy of ours is higher. With the proposed approach, we give an efficient support to the LBSs provider in monitoring user intelligently and sending information to user in a push-driven fashion. Apart from the support of timely and desired services and enhanced automation, the technique helps overcome some existing issues such as network flooding due to the massive tracking of users, the latencies of the positioning systems in providing and information delivery. Accordingly, the positioning is more reliable, which enables the service provider to effectively and efficiently offer location-based services with high frequency. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction The constant advances in hardware technologies have given rise to a personal communication system, which ensure the availability of selective information disseminating services and LBSs (Stojanov- ic & Djordjevic-Kajan, 2001). Such hardware trends include devel- opments in miniaturization of electronics technologies, display devices, and wireless communications. Other trends are the im- proved performance of general computing technologies and the general improvement in performance of electronic hardware. And especially positioning technologies such as Global Positioning Sys- tem (GPS) are becoming increasingly accurate. Up to now LBSs sys- tem have been known as a distributed mobile computing environment where the geographic locations of users in space are particularly utilized for computing and application-related optimization besides data on the past interactions with the users, data accessible through the Internet and other data obtain from sensors. Examples of LBSs include location-aware advertising, traf- fic coordination, management, any way finding, integrated infor- mation services, e.g., tourist services (Cheverst, Davies, Mitchell, & Friday, 2000), a nearest available parking lot application (Chon, Agrawal, & Abbadi, 2002), and a LBS framework using Cellular Dig- ital Packet Data (Jana, Johnson, Muthukrishnan, & Vitaletti, 2001). Current LBSs use information about users’ current locations to determine services which the users request. Data derived from user requests are integrated with other user data in a multidimen- sional data model for later analysis to customize the users’ context and demand (Jensen, Kligys, Pedersen, & Timko, 2004). Although the common computing strategy in LBSs benefits the users, there are additional benefits when future locations are predicted. One major advantage of location prediction is that it provides LBSs with extended resources, mainly time, to improve system reliability which in turn increases the users’ confidence and the demand for the services. In other words, location prediction provides a longer time available to prepare and present services, in particular ser- vices involving complex and time consuming task and to ensure that only desired services are delivered. Having prior information about locations where the user will visit at later times during a trip will extend the location management capability of LBSs and will facilitate the generation of new services that were impossible 0360-8352/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.cie.2008.07.009 * Corresponding author. Tel.: +1 740 597 3144. E-mail addresses: nhanvth@etri.re.kr (T.H.N. Vu), khryu@dblab.chungbuk.ac.kr (K.H. Ryu), parkn@ohio.edu (N. Park). Computers & Industrial Engineering 57 (2009) 91–105 Contents lists available at ScienceDirect Computers & Industrial Engineering journal homepage: www.elsevier.com/locate/caie