作者: Guochen Cai , Kyungmi Lee , Ickjai Lee
DOI: 10.1016/J.ESWA.2017.10.049
关键词: Trajectory pattern 、 Semantics 、 Collaborative filtering 、 Set (abstract data type) 、 Computer science 、 Recommender system 、 World Wide Web
摘要: Abstract A large number of geo-tagged photos become available online due to the advances in geo-tagging services and Web technologies. These are indicative photo-takers’ trails movements, have been used for mining people movements trajectory patterns. inherently spatio-temporal, sequential implicitly containing aspatial semantics. recommender systems collaborative filtering based. There some studies build itinerary from these photos, but they fail consider dimensions share common drawbacks, especially lacking semantics or temporal information. This paper proposes an system with semantic pattern by discovering points-of-interest information other users’ visiting sequences preferences. Our considers sequential, dimensions, also takes into account user-specified preferences constraints customise their requests. It generates a set customised targeted semantic-level itineraries meeting user specified constraints. The proposed method historic people’s frequent travel patterns photos. Experimental results demonstrate informativeness, efficiency effectiveness our over traditional approaches.