Abstract

Nowaday, the rapid popularization of mobile terminals and increasing maturity of location technology are bringing new opportunities and challenges to recommendation systems. Recommendation systems need to consider user's mobility and the resulting related issues, such as the relationship between user's interests and locations, and the requirement of real-time. The problems which facing are more complex. Besides, if making good use of user's location information, can improve the performance of recommendation. In this paper, an outdoor recommendation algorithm based on user's location is proposed. When user is in a different position, using different recommendation algorithms to generate recommendations, it overcomes the cold start and sparse matrix problem which in traditional recommendation algorithms, and it also can alleviate the problem of capturing user's short-term interests in mobile environment. The experimental results show that the proposed algorithm is better than other traditional recommendation algorithms.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call