| Summary: | As the use of smart mobile devices increases, users can connect and coordinate with local people or events that match their interests through geosocial networking applications. Such applications require techniques that support multiple-user location-based keyword (MULK) query processing, which takes a set of geo-locations and a set of keywords as arguments and returns a list of relevant POIs that match the arguments. In geosocial networking, processing this query involves two issues: mining users' individual preferences and indexing road networks. In this paper, we investigate how to mine users' individual preferences by enhancing MULK query processing with interactions, resulting in interactive multiple-user location-based keyword query. Furthermore, we propose a hierarchical road network index structure to support efficient query processing. Finally, we conduct a set of experiments that demonstrate the effectiveness and efficiency of our solutions under various parameter settings.
|
|---|
