|Ph.D Student||Mor Matan|
|Subject||Exploring Travel Activities of Different User Groups|
through Social Media Photos for Customized Route
|Department||Department of Civil and Environmental Engineering||Supervisor||PROF. Sagi Dalyot|
Geotagged photos are uploaded by users to social media photo-sharing online websites, which become more popular and commonly used by travelers to share their experiences. Handling, mining, and interpreting these user-generated ‘digital footprints’ can be used to reconstruct travel trajectories of users to recover their activity and knowledge on the urban environment. In this research, Flickr geotagged crowdsource photo database is showcased as a source for mining users’ trajectories to effectively explore different travel activities, later used to plan customized walking routes. Two different types of photo contributors, namely tourists and locals, are distinguishing using two classification algorithms: 1) adaptive spatiotemporal descriptors, and 2) machine learning. These are used to develop two matrices: popular places and popular connectivity (transition) between them, which should resemble the overall travel activity patterns of both user groups. The matrices are used as a context graph in two routing approaches: 1) greedy, and 2) Orienteering Problem. Several examples are presented with various trade-offs between different route objectives that produce customized routes based on user preferences. The results of this research show that with adapted data handling of geotagged user-generated content, implicit travel information can be explored for measuring and analyzing travel activity of different groups to enrich existing knowledge related to travel analysis and management in urban areas.