Efficient Probabilistic Reverse k-Nearest Neighbors Query Processing on Uncertain Data

摘要: A reverse k-nearest neighbors (RkNN) query returns all the objects that take object q as their k nearest neighbors. However, data are often uncertain in numerous applications. In this paper, we focus on problem of processing RkNN data. probabilistic (PRkNN) retrieves have higher probabilities than a user-specified threshold to be q. The previous work for answering PRNN mainly based distance relationship between objects, and inapplicable PRkNN when > 1. design novel algorithm support arbitrary values basis two pruning strategies, namely spatial pruning. rule is defined both distances angle ranges objects. And an efficient upper bound probability estimated by algorithm. Extensive experiments conducted study performance proposed approach. results show our has better scalability existing solution regarding growth k.