IEICE Transactions on Information and Systems
Online ISSN : 1745-1361
Print ISSN : 0916-8532
Regular Section
Hybrid Markov Location Prediction Algorithm Based on Dynamic Social Ties
Wen LIShi-xiong XIAFeng LIULei ZHANG
Author information
JOURNAL FREE ACCESS

2015 Volume E98.D Issue 8 Pages 1456-1464

Details
Abstract

Much research which has shown the usage of social ties could improve the location predictive performance, but as the strength of social ties is varying constantly with time, using the movement data of user's close friends at different times could obtain a better predictive performance. A hybrid Markov location prediction algorithm based on dynamic social ties is presented. The time is divided by the absolute time (week) to mine the long-term changing trend of users' social ties, and then the movements of each week are projected to the workdays and weekends to find the changes of the social circle in different time slices. The segmented friends' movements are compared to the history of the user with our modified cross-sample entropy to discover the individuals who have the relatively high similarity with the user in different time intervals. Finally, the user's historical movement data and his friends' movements at different times which are assigned with the similarity weights are combined to build the hybrid Markov model. The experiments based on a real location-based social network dataset show the hybrid Markov location prediction algorithm could improve 15% predictive accuracy compared with the location prediction algorithms that consider the global strength of social ties.

Content from these authors
© 2015 The Institute of Electronics, Information and Communication Engineers
Previous article Next article
feedback
Top