Research On Group Reverse Farthest Neighbor Query Technology

With the rapid development of intelligent transportation and geographic information system,spatial database query technology has attracted the attention of many scholars.Among them,the reverse furthest neighbor query from the data point to find the target point as its farthest adjacent data points,used to obtain target set of weak influence,its research results has been widely applied to facility location,earthquake relief,marketing and other major areas,thus,reverse furthest neighbors query technology research is of great significance and practical value.However,in view of the multiple target point in the practical situation of reverse furthest neighbor query problem,the existing research results has certain limitation,therefore,this article puts forward the European space group reverse furthest neighbor query and obstacle space group reverse k furthest neighbors query these two kinds of problems,and carried on the thorough discussion,the main research content is as follows.First,the concept of the Farthest neighbor query based on Group Reverse in Euclidean space and the algorithm of the Farthest neighbor query based on the smallest covered Circle(C-GRFN)are proposed.The algorithm firstly obtains the minimum coverage circle and its center of the query point set,and considers all the query point sets as a whole to reduce the access to the query point set.Then,the data point set is filtered by the pruning strategy based on the quartile neighborhood and p-ray.Finally,the reverse range query algorithm is used to refine the data set,reduce the data set,and obtain the final result.Secondly,the concept of group reverse farthest neighbor query is extended to obstacle space,and the concept of group reverse k farthest neighbor query in obstacle space is proposed,and a query optimization algorithm based on Voronoi graph(Voronoi-Obstacle Group Reverse k Farther-Neighbors,V-OGRkFN)is given,this algorithm firstly solve the query point set minimum cover round query algorithm for group to simplify the problem,build Voronoi diagram index and logarithmic stronghold set,by pruning strategy based on Voronoi diagram properties to filter data points,and the candidate set is obtained by virtue of the transformation relation between reverse-k nearest neighbors and reverse-k farthest neighbors.Finally,the point set of candidate set is purified by refining algorithm,which improves the accuracy of query and solves the problem of group reverse k-farthest neighbor query in obstacle space.Finally,the above algorithms are verified by experiments.Among them,the group reverse farthest neighbor query algorithm in Euclidean space designs a comparative experiment from the aspects of data set capacity,query point set size and spatial distribution of data.The reverse k-farthest neighbor query algorithm for obstacle space group designs a comparative experiment from the perspectives of data set capacity,query point set size,obstacle set size,k value and obstacle rate.Experimental results show that the proposed algorithm has good query performance.