摘要
数据流的模式查询具有很高的领域价值,它不仅需要较高的抗噪能力和实时性,而且查询目标模式还具有可伸缩性,即由多个子模式复合而成,且某些子模式可重复、缺失或倒置.文中提出一种可伸缩模式的查询(SPQ)方法,允许用户定义目标模式并设置可伸缩条件.然后在查询处理中通过模式匹配生成模式流,进而完成满足可伸缩条件的目标模式查询.在真实数据集上进行的实验从查全率、查准率和处理效率上证明了SPQ方法是可行和灵活的.
Pattern query over data streams possess high domain significance.It requires anti-noise capability and real time processing.Meanwhile,in many cases,the query target pattern is also scalable,which means it is comprised of sub-patterns,with some sub-patterns gained,lost or even inverse.This paper presents a scalable pattern query（SPQ） method.It allows users to define target pattern and set corresponding scalable constraints according to their knowledge and concerns,finally the target pattern can be changed to scalable pattern.In the stage of query evaluation,pattern stream is generated by pattern matching,and then scalable pattern query is carried out on the pattern stream.From the perspective of recall ratio,precision ratio and processing efficiency,the experimental results on real datasets show that SPQ is feasible and flexible.
出处
《计算机学报》
EI
CSCD
北大核心
2010年第8期1481-1491,共11页
Chinese Journal of Computers
基金
国家自然科学基金(60673113
60973002)
国家"八六三"高技术研究发展计划项目基金(2007AA01Z191
2009AA01Z150)资助~~
关键词
数据流
查询
可伸缩模式
目标模式
查询重写
查询处理
data stream
query
scalable pattern
target pattern
query rewrite
query evaluation
