In recent years, the nearest neighbor search （NNS） problem has been widely used in various interesting applications. Locality-sensitive hashing （LSH）, a popular algorithm for the approximate nearest neighbor probl...In recent years, the nearest neighbor search （NNS） problem has been widely used in various interesting applications. Locality-sensitive hashing （LSH）, a popular algorithm for the approximate nearest neighbor problem, is proved to be an efficient method to solve the NNS problem in the high-dimensional and large-scale databases. Based on the scheme of p-stable LSH, this paper introduces a novel improvement algorithm called randomness-based locality-sensitive hashing （RLSH） based on p-stable LSH. Our proposed algorithm modifies the query strategy that it randomly selects a certain hash table to project the query point instead of mapping the query point into all hash tables in the period of the nearest neighbor query and reconstructs the candidate points for finding the nearest neighbors. This improvement strategy ensures that RLSH spends less time searching for the nearest neighbors than the p-stable LSH algorithm to keep a high recall. Besides, this strategy is proved to promote the diversity of the candidate points even with fewer hash tables. Experiments are executed on the synthetic dataset and open dataset. The results show that our method can cost less time consumption and less space requirements than the p-stable LSH while balancing the same recall.展开更多
Reverse k nearest neighbor （RNNk） is a generalization of the reverse nearest neighbor problem and receives increasing attention recently in the spatial data index and query. RNNk query is to retrieve all the data po...Reverse k nearest neighbor （RNNk） is a generalization of the reverse nearest neighbor problem and receives increasing attention recently in the spatial data index and query. RNNk query is to retrieve all the data points which use a query point as one of their k nearest neighbors. To answer the RNNk of queries efficiently, the properties of the Voronoi cell and the space-dividing regions are applied. The RNNk of the given point can be found without computing its nearest neighbors every time by using the rank Voronoi cell. With the elementary RNNk query result, the candidate data points of reverse nearest neighbors can he further limited by the approximation with sweepline and the partial extension of query region Q. The approximate minimum average distance （AMAD） can be calculated by the approximate RNNk without the restriction of k. Experimental results indicate the efficiency and the effectiveness of the algorithm and the approximate method in three varied data distribution spaces. The approximate query and the calculation method with the high precision and the accurate recall are obtained by filtrating data and pruning the search space.展开更多
基金Project supported by the National Natural Science Foundation of China(Grant No.61173143)the Special Public Sector Research Program of China(Grant No.GYHY201206030)the Deanship of Scientific Research at King Saud University for funding this work through research group No.RGP-VPP-264
文摘In recent years, the nearest neighbor search （NNS） problem has been widely used in various interesting applications. Locality-sensitive hashing （LSH）, a popular algorithm for the approximate nearest neighbor problem, is proved to be an efficient method to solve the NNS problem in the high-dimensional and large-scale databases. Based on the scheme of p-stable LSH, this paper introduces a novel improvement algorithm called randomness-based locality-sensitive hashing （RLSH） based on p-stable LSH. Our proposed algorithm modifies the query strategy that it randomly selects a certain hash table to project the query point instead of mapping the query point into all hash tables in the period of the nearest neighbor query and reconstructs the candidate points for finding the nearest neighbors. This improvement strategy ensures that RLSH spends less time searching for the nearest neighbors than the p-stable LSH algorithm to keep a high recall. Besides, this strategy is proved to promote the diversity of the candidate points even with fewer hash tables. Experiments are executed on the synthetic dataset and open dataset. The results show that our method can cost less time consumption and less space requirements than the p-stable LSH while balancing the same recall.
基金Supported by the National Natural Science Foundation of China (60673136)the Natural Science Foundation of Heilongjiang Province of China (F200601)~~
文摘Reverse k nearest neighbor （RNNk） is a generalization of the reverse nearest neighbor problem and receives increasing attention recently in the spatial data index and query. RNNk query is to retrieve all the data points which use a query point as one of their k nearest neighbors. To answer the RNNk of queries efficiently, the properties of the Voronoi cell and the space-dividing regions are applied. The RNNk of the given point can be found without computing its nearest neighbors every time by using the rank Voronoi cell. With the elementary RNNk query result, the candidate data points of reverse nearest neighbors can he further limited by the approximation with sweepline and the partial extension of query region Q. The approximate minimum average distance （AMAD） can be calculated by the approximate RNNk without the restriction of k. Experimental results indicate the efficiency and the effectiveness of the algorithm and the approximate method in three varied data distribution spaces. The approximate query and the calculation method with the high precision and the accurate recall are obtained by filtrating data and pruning the search space.
