Multi-user Skyline Interaction Query Processing

In the era of big data,the data is multi-dimensional and sea-quantified.The query tends to be personalized,and users have higher requirements for accuracy and efficiency.The traditional multi-dimensional data query method can not effectively solve the problem of combining static and spatial attributes,multi-user query of dynamic space attributes,and personalized query on the categorical domain.Therefore,research on multi-user queries in the metric space and obtain user preference on the categorical domain are very important for finding high quality result sets.Multi-dimensional Skyline queries are mainly divided into two categories based on the attribute value characteristics of object data points: numeric fields and categorical fields.This thesis focuses on the problem of spatial numerical attribute query and user interaction of categorical domain attribute,and designs and proposes an efficient multi-user Skyline query interaction algorithm.The main research work is as follows:(1)Most of the Skyline queries in European space only explore the spatial properties,but ignore the static attributes of data points.Based on the properties of Voronoi graph,R-tree and convex hull,this thesis proposes the concept of search region and gives Voronoi R-tree Search algorithm.The algorithm effectively reduces the number of dominance tests between data points and distance calculation by pruning the data set and staged dominant tests.The effectiveness of the algorithm is verified in real and simulated data sets.(2)Skyline query only consider the distance in the dynamic attribute,but the effect of user speed on query results is not considered.This thesis presents a query algorithm Exit In,which can deal with the change of user's speed.According to the nature of segmentation points,the algorithm can update Skyline result set in real time according to user's status by using the association among multiple queries.The effectiveness of the proposed algorithm is verified through a large number of experiments on the real road network,and compared with the existing algorithms.(3)Most of the existing algorithms perform query processing on the numerical domain.When applied to the categorical domain attribute,the result set size is uncontrollable.To this end,the Skyline query problem on the categorical domain is proposed.The dominating table and the matrix index which can describe the dominance relationship of the data points are given,and the selection strategy of the attribute value pairs is proposed based on the index,and the efficient selection and elimination of the attribute value pairs are realized.The experimental results compared with the existing algorithms show that the Skyline Matrix Index Interactive algorithm exhibits higher query efficiency under different parameters.