ABSTRACT
Social friendship and geographical position information often reflects individuals’ personal preferences and other types of knowledge that can be used to extract their similarity for recommendation systems. This paper finds that users are more likely to move around some specific centres or check in at some hotspots; a few individuals check in frequently, whereas most locations are rarely visited. Based on these findings, we propose a multi-centre clustering algorithm to capture users’ mobile patterns and develop a user similarity measurement method. Complexity analysis shows the method’s efficiency in handling large datasets and experimental results demonstrate its good applicability.
Acknowledgments
The study is partly support by the Zhejiang Federation of Humanities and Social Sciences (No. 19ZJQN20YB), the Natural Science Foundation of Ningbo (project title: Study on Trust Mining in Shared Travel Domain), the Ningbo Academy of Social Sciences (project title: the Role and Development Strategy of Cross-border E-commerce in Ningbo’s Transition from a Big Foreign Trade City to a Strong Foreign Trade City), the Key Laboratory of Geographic Information Science (Ministry of Education), East China Normal University (No. KLGIS2014A0l) and the Modern Port Service Industry and Culture Research Center of the Key Research Base of Philosophy and Social Sciences of Zhejiang Province.
Disclosure statement
No potential conflict of interest was reported by the authors.