Algorithm for Fast Searching of k-nearest Neighbors in Cloud Points
Lijuan Wu · Mini-micro Systems · 2007
A new algorithm of fast searching k nearest neighbors in cloud data is presented in the paper A method of estimating the size of sub-cube is introduced by considering the range of data set , the total numbers , the density of points and the numbers of nearest neighbors. Then the minimal box of the data set is divided into a set of sub-cubes and the searching speed is decided by the number of the sub-cubes. Finally, the points are assigned to the appropriate sub-cubes for searching k nearest neighbors of the testing point. Simulation experiments show that the searching speed of k nearest neighbors is improved.