Research On A Skyline Query In Three-dimensional Obstacle Space

Skyline query,as a query to solve typical multi-objective optimization problems,has a wide range of applications in market analysis and data mining.Since its proposal,it has attracted the attention of researchers,and has developed into the later spatial Skyline query,probabilistic Skyline query,and Top-k Skyline query.However,when users need to consider the non-spatial and spatial attributes of the data set at the same time,and there are differences in the importance of each attribute,the traditional Skyline query cannot obtain the correct result set.To solve these problems,a new kind of Skyline-like query is proposed.Then considering the application of the proposed new Skyline query in different scenarios,the solutions of 2D Euclidean space and 3D obstacle space are given.This paper conducts a systematic research on Skyline-like query.The main research contents are as follows.First,a Skyline-like query in a two-dimensional European space is proposed.Mainly introduce the research background and problem definition of the query.In order to solve the query problem,based on the traditional Skyline query,the concept of dominating domain is proposed and the detail algorithm of Skyline-like query is given,The Skyline-like points in the data set are obtained,Finally the result set is sorted from high to low according to the user's importance of each attribute,and the result set is returned.Secondly,the obstacle Skyline query in the three-dimensional obstacle space is proposed.The main consideration is that the query is different from the Skyline-like query when obtaining a result set with advantageous spatial attributes.Principally,the influence of obstacles is considered when calculating the distance.In the process of calculating the obstacle distance,in order to improve efficiency,the obstacles are pruned in advance and the R-tree nodes storing the data points are filtered.Then a three-dimensional view is constructed,and an example is given to illustrate its correctness.Besides,the algorithm implementation and detailed algorithm analysis of the Skyline query of 3D obstacles are given.Finally,comparative experiments are carried out on the two Skyline-like query algorithms proposed above.First,the Skyline-like query is compared with the B~2S~2algorithm.The experimental results show that the Skyline-like algorithm is better than the traditional algorithm in terms of query time.Then 3D obstacle Skyline queries conduct internal comparison experiments by seting up data sets of different sizes and different numbers of obstacles.According to the experimental results,the influence on query efficiency is analyzed.