Recommending Locations Based on Users’ Periodic Behaviors

Abstract EN

The research of location recommendation system is an important topic in the field of LBSN (Location-Based Social Network).

Recently, more and more researchers began focusing on researching how to recommend locations based on user’s life behavior.

In this paper, we proposed a new model recommending locations based on user’s periodic behaviors.

In view of multiple periodic behaviors existing in time series, an algorithm which can mine all periods in time series is proposed in this paper.

Based on the periodic behaviors, we recommend locations using item-based collaborative filtering algorithm.

In this paper, we will also introduce our recommendation system which can collect users’ GPS trajectory, mine user’s multiple periods, and recommend locations based user’s periodic behavior.