Diversifying Results of Top-k Querys over Bounded Regions

Existing solutions of top-k query processing require traversing over all objects in initial result set. To improve the efficiency of query processing, we map initial result set to the Euclidean space and select diversification optimal subspace using two methods according to preferential strategy. Finally, we propose top-k bounded diversification algorithm. The algorithm ensures that the results of the second query with similarity and difference. Experimental results show that the proposed method significantly reduces the number of accessing to the object.