Recommending Locations Based on Users’ Periodic Behaviors

Abstract

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.

A Cold Start Context-Aware Recommender System for Tour Planning Using Artificial Neural Network and Case Based Reasoning

“…Tourist plans maximize the user utility and POIs travel distance. Reference [28] mines multiple periods in the time sequence to calculate the user's period of arriving at a certain area. Then they use item-based collaborative filtering for recommending locations based on users' preference and the periodic behaviors.…”

“…However, and to the best of our knowledge, previous context-aware tourism recommender systems developed so far have not completely addressed these issues [9,[11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28]. One weakness of these researches is that they often suggest to a given user a list of POIs, which are specific tourist attractions that a user may find interesting such as a historical or cultural buildings or any places worth being visited, rather than a tour as an ordered set of POIs.…”