Travel Route Recommendation Using User Mobility and Social Interactions
The popularity of social media (for example, Face Book and Flicker) makes it easy for users to share their check-in documents and images on their journey. In order to promote the preparation of travel, we want to explore the vast amount of user history records in social media. Users always have certain preferences regarding their trips when planning a trip. We accept arbitrary text explanations as keywords regarding customized queries, rather than limiting users to restricted query choices such as pages, events or time periods. In addition, a range of recommended routes are necessary and representative. Previous work on the mining and classification of existing routes for check-in data has been developed. We claim that more features of Places of Interest should be removed in order to meet the necessity of an automatic tour organization. Therefore we are proposing in this paper an efficient framework that used knowledge extraction from historical user mobility and social interactions, with keyword-aware representative travel route. In order to effectively match query keyword, we specifically have designed the keyword extraction module for classifying the POI tags. We have further developed an algorithm for road reconstruction to build road candidates who satisfy the requirements. We examine Representative Skyline Concepts, i.e. the Skyline roads, to provide appropriate query results, which describes the best compromise between various POI features. In order to test the efficiency and efficacy of the proposed algorithms, we have carried out comprehensive experiments with real-locality social network sets. Experimental findings indicate that our approaches are indeed good when comparing to cutting-edge works.