Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
Mining interesting locations and travel sequences from GPS trajectories
Proceedings of the 18th international conference on World wide web
| Hi-index | 0.00 |
Nowadays, due to the increasing user requirements of efficient and personalized services, a perfect travel plan is urgently needed. However, at present it is hard for people to make a personalized traveling plan. Most of them follow other people's general travel trajectory. So only after finishing their travel, do they know which scene is their favorite, which is not, and what is the perfect order of visits. In this research we propose a novel spatio-temporal sequence (STS) searching, which mainly includes two steps. Firstly, we propose a novel method to detect tourist features of every scene, and its difference in different seasons. Secondly, combined with personal profile and scene features, a set of interesting scenes will be chosen and each scene has a specific weight for each user. The goal of our research is to provide the traveler with the STS, which passes through as many chosen scenes as possible with the maximum weight and the minimum distance within his travel time. We propose a method based on topic model to detect scene features, and provide two approximate algorithms to mine STS: a local optimization algorithm and a global optimization algorithm. System evaluations have been conducted and the performance results show the efficiency.