Classifying urban events' popularity by analyzing friends information in location-based social network

Abstract

Recent progress and spread of smartphones and social network services have enabled us to transmit text messages with GPS location data anywhere and anytime. Since these location-based SNS messages often refer to urban events, many researchers have tried to recognize urban events by analyzing of the messages. To construct the various applications based on the urban events information, we propose a new indicator of event, called Popularity which represents how popular the urban event is. Popularity is estimated by analyzing friends on social network of events' participants. To evaluate our new indicator, we designed and implemented intuitive and interactive web-based tool for analyzing Popularity of events. Through comparative experiments, we confirmed that our proposed method could provide a certain amount of accuracy for estimating Popularity of events.

Publication
URB-IOT ‘14: Proceedings of the First International Conference on IoT in Urban SpaceOctober 2014 Pages 87–89
Makoto Kawano
Researcher

My research interests include efficient machine learning for real-world deployment.