The optimal-location query

摘要: We propose and solve the optimal-location query in spatial databases. Given a set S of sites, O weighted objects, region Q, returns location Q with maximum influence. Here influence l is total weight its RNNs, i.e. objects that are closer to than any site S. This new has practical applications, but very challenging solve. Existing work on computing RNNs assumes single location, thus cannot be used compute optimal locations. The reason there infinite candidate locations Q. If we check finite locations, result can inaccurate, revealed may not have paper proposes three methods accurately first method uses standard R*-tree. To an retrieves certain from R*-tree sends them as stream plane-sweep algorithm, which data structure called aSB-tree ensure efficiency. second based index OL-tree, novelly extends k-d-B-tree store segmented rectangular records. OL-tree only theoretical usage for it space efficient. most approach Virtual OL-tree. These theoretically experimentally evaluated.