Font Size: a A A

Research And Implementation Of Skyline Query Algorithm In LBSN Environment

Posted on:2021-02-05Degree:MasterType:Thesis
Country:ChinaCandidate:F Y SongFull Text:PDF
GTID:2518306329484244Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the popularity of smart mobile terminals,location-based social network(LBSN)has developed rapidly.As an important query in the research field of LBSN,skyline query plays an important role in location-based social network platform,financial data analysis and medical data analysis.The purpose of this query is to find representative objects from the data set and recommend them to the users and facilitate the users to make decisions.However,an important problem is that the size of skyline set is sensitive to data distribution,dimension and scale.In addition,with the generation of a large amount of text data,skyline query combined with text information,that is,keyword skyline query has become more and more important.However,the skyline query for text data under streaming data has not been studied.To address the above problems,two algorithm frameworks,PAKRS(Predict-based Representative Skyline)and PCKSWI(Partition-based Continuous Keyword Skyline with Index),are proposed in this paper.In this paper,we first propose a p-approximate k representative skyline query.The query returns the approximate k representative skyline set in the current window,and ensures that the ratio of the dominance size of the approximate k representative skyline set and the exact query result set does not exceed threshold value ?.In order to support this query,we propose a novel framework named PAKRS(Predict-based Approximate k Representative Skyline).Firstly,PAKRS partitions the current window into a group of sub-windows.Next,we construct the predicted result sets of a few future windows according to the partition results.In this way,we can predict the moment that when new arrived objects can become skyline object.Secondly,to solve the problem of continuous keyword skyline query over sliding window,this paper proposes a keyword skyline query algorithm over streaming data.Different from the traditional skyline query,it computes the skyline set for the objects matching the given keyword and returns the query result to the user.By partitioning the current window,the algorithm filters out most objects that cannot be the query result,and overcomes the influence of timing relation between data on the performance of the algorithm.After that,this paper proposes the index KSG(Keyword Skyline Grid)to realize the effective keyword filtering and skyline filtering for the objects in each partition.In addition,by adjusting the granularity of partition,this paper realizes the effectively partition of window and improve the query efficiency.Finally,this paper evaluates the performance of the proposed algorithm through a large number of experiments,and implements a recommendation system.The experimental results prove the effectiveness and high efficiency of the proposed algorithm.
Keywords/Search Tags:Skyline Query, Keyword Skyline Query, k Representative Skyline Query, Streaming Data, Sliding Window
PDF Full Text Request
Related items