数据流上的连续预测聚集查询  

Predictive Continuous Aggregate Queries over Data Streams

作　　者：郭龙江[1] 李建中[1] 王伟平[1] 张冬冬[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院

出　　处：《计算机研究与发展》2004年第10期1690-1695,共6页Journal of Computer Research and Development

基　　金：国家自然科学基金重点项目 ( 60 2 73 0 82 ) ;黑龙江省自然科学基金重点项目 (ZJG0 3 0 5 )

摘　　要：提出了一种数据流上未来值的连续查询 ,称为连续预测查询 采用数理统计的方法给出了带有COUNT聚集函数的连续预测聚集查询实现算法 通过采用TPC H标准测试数据和随机生成的模拟数据进行了实验 理论和实验结果表明 。Predictive continuous aggregate queries over data streams are proposed in this paper Models and algorithms for predicting the aggregate value, such as count, over data streams based on recent data are given using statistics Extensive experimental studies based on real benchmark data and synthetic data are carried out to demonstrate the accuracy of the techniques Analysis and experimental results show that the techniques in terms of statistics could obtain better accuracy of prediction of aggregate values over data streams in the future

关 键 词：数据流 连续查询 连续预测查询 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]