Abstract: Nowadays,due to the increasing user requirements of efficient and personalized services,a perfect trip planning is urgently needed.However,it is hard for people to make a personalized traveling plan at present.For a satisfactory trip planning,the following features are desired:i) personalized recommender based on the interest and habits of travelers；ii) maximal coverage of sites of interest；iii) minimal effort such as transporting time on the route.As a trip contains a sequence of stops at multiple scenes,the problem of trip planning becomes an optimized spatio-temporal sequence where each stop is weighted.For each city,this paper generated a set of weighted scenes for each user.Then it can retrieve the optimal sequence of scenes in terms of distance,weight,visiting time and scene features.We developed alternative algorithms for searching optimal sequences,with consideration of the weight of each scene,the preference of users,and the travel time constraint.The experiments demonstrate the efficiency of the proposed algorithms based on real datasets from social networks.

Key words: Spatio-temporal,Sequence,Social networks,Personalized,Trip planning