| Recent years,with the booming of mobile Internet and the global positioning technology,a massive amount of spatial-textual data are continuously generated from web and applications of mobile phones.The dissemination and delivering of those data become more and more important.Location-aware subscribe/publish is a hot topic in message dissemination area.Thus,it aroused wide concern both in academics and industries.This paper researched subscription matching in location-aware subscribe/publish and designed a series of efficient algorithms for subscription matching over data type of Boolean expressions and big data stream.The main researches in this paper are as follows:(1)Analyze background significance and related research status at location-awar e publish/subscribe and then discuss the main ideas of traditional location-aware sub scribe/publish with advantages and shortcomings to support the following research w ork.(2)Focus on problem of matching problem in subscription with boolean expressions in location-aware subscribe/publish,we proposed an index and querying solutions named RP-trees.This solution combined some features of Opindex and R-tree and can efficiently prune both in boolean expressions and spatial information dimension which can meet the requirements for matching problem in subscription with boolean expressions.(3)To resolve top-k boolean querying and matching problem in location-aware subscribe/publish.We proposed an index and query solutions named RR~t-trees.This solution utilize the partial features of R~t-tree and Opindex and combined the algorithm of top-k boolean subscription querying and matching.This solution can maintain a result set of top-k subscriptions for each event,which effectively resolved the top-k boolean querying and matching problem.Finally,we compared our solution with baseline solutions by conducting experiments both on synthetic and real-world datasets and realized the visualization demo system for location-aware subscribe/publish. |
PDF Full Text Request