Abstract
The rapidly growing of Location-based Social Networks (LBSNs) provides a vast amount of check-in data, which enables many services, e.g., point-of-interest (POI) recommendation. In this paper, we study the next new POI recommendation problem in which new POIs relating to users’ current location are to be recommended.
The problem complexity lies in the difficulty in precisely learning users’ sequential information and personalizing the recommendation model. To this end, we resort to the Metric Embedding method that avoids drawbacks of the commonly used Matrix Factorization technique in the recommendation.
We propose a personalized ranking metric embedding method (PRME) to model personalized check-in sequences. We further develop a PRME-G model, which integrates sequential information, individual preference and geographical influence, to improve the recommendation performance. Experiments on two real-world LBSNs datasets demonstrate that our new algorithm outperforms the state-of-the-art next POI recommendation methods.
The problem complexity lies in the difficulty in precisely learning users’ sequential information and personalizing the recommendation model. To this end, we resort to the Metric Embedding method that avoids drawbacks of the commonly used Matrix Factorization technique in the recommendation.
We propose a personalized ranking metric embedding method (PRME) to model personalized check-in sequences. We further develop a PRME-G model, which integrates sequential information, individual preference and geographical influence, to improve the recommendation performance. Experiments on two real-world LBSNs datasets demonstrate that our new algorithm outperforms the state-of-the-art next POI recommendation methods.
| Original language | English |
|---|---|
| Title of host publication | IJCAI'15 Proceedings of the 24th International Conference on Artificial Intelligence |
| Publisher | ACM |
| Pages | 2069-2075 |
| ISBN (Electronic) | 978-1-57735-738-4 |
| Publication status | Published - 2015 |