空间文本数据流上连续查询评估技术综述  

Survey of Continuous Queries over Spatial-Textual Data Streams

作　　者：杨茸 牛保宁 YANG Rong;NIU Baoning(College of Information and Computer,Taiyuan University of Technology,Jinzhong,Shanxi 030600,China)

机构地区：[1]太原理工大学信息与计算机学院,山西晋中030600

出　　处：《计算机科学与探索》2021年第4期631-640,共10页Journal of Frontiers of Computer Science and Technology

基　　金：国家自然科学基金(61572345);国家重点研发计划(2017YFB1401000)。

摘　　要：空间文本数据流上连续查询(CQST)在基于位置的服务中应用广泛,其在不断更新的数据流上,持续监控满足空间和文本约束的结果。为了将数据流中的对象尽快匹配给CQST,在CQST上构建高效的过滤技术是关键。CQST查询评估方法——为查询选取恰当的空间文本索引,构建高效的过滤策略提升索引的空间文本过滤性能,为数据流中到来的对象过滤大量不相关的查询,避免高昂的验证代价,提高对象与查询的匹配效率。现有工作利用有限的空间索引和文本索引构建空间文本混合索引,其评估性能差异取决于采用的过滤策略,即提升索引过滤性能的技术。以现有CQST查询优化技术为主要研究对象,对评估CQST的流程以及存在的挑战进行了介绍;对在中央服务器及分布式集群上评估CQST的空间过滤技术及文本过滤技术进行综述比较,包含采用的空间文本混合索引,为提升索引过滤性能采用的空间过滤策略、文本过滤策略及二者的结合机制,分析总结其利弊,讨论评估CQST未来可能的研究方向。The continuous queries over spatial-textual data streams(CQST)are widely used in location-based services,which continuously monitor the results satisfying spatial and textual constraints over updated data streams.To match the objects over the data streams with CQST as soon as possible,building efficient filtering techniques on the CQST is the key.The approach to evaluating CQST is to select the appropriate spatial and textual index to organize the CQST,construct efficient filtering strategies to improve the spatial and textual filtering capabilities of the index,filter large number of unpromising queries for the objects over the data streams,avoid costly verification cost and improve the efficiency of matching objects and queries.The existing works construct spatial-textual hybrid index by using limited spatial index and textual index,and the difference in evaluation performance depends on the filtering strategies used,i.e.the techniques of improving the filtering performance of the index.This paper takes the techniques of evaluating CQST as the research object,introduces the framework and challenges of evaluating CQST,reviews and compares the spatial and textual filtering techniques of evaluating CQST on a central server and distributed clusters,including the adopted spatial-textual hybrid index,the spatial and textual filtering strategies and the combination scheme of the spatial and textual index to improve the filtering performance of the index,analyzes and summarizes their advantages and disadvantages,and discusses possible future research directions.

关 键 词：空间文本查询 连续查询 过滤策略 数据流 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]