摘要
设计了数据流预测查询的新模型,包括局域流能量预测、能量分布模式挖掘及预测序列的重构和数据流能量的度量方法;设计了融合数据流能量回归与基于频繁模式的小波分解预测新方法,并将新算法推广到强偶合多数据流的预测查询;提出了最近最频繁序列模式的新概念,并应用于局域流能量分解;在真实数据上的模拟实验,验证了算法的有效性.
A new predict model was contrived, which involves local stream energy prediction, the energy distribution pattern mining, the predictive series reconstruction and measurement method of stream energy. A new method was designed to forecast stream by energy regression and wavelets decomposing based on frequent pattern, and extended to multi-streams with strong coincidence. The concept of the nearest maximum frequent pattern was proposed to decompose local stream energy. The validity of new algorithm was demonstrated by extensive experiments on real data.
出处
《软件学报》
EI
CSCD
北大核心
2008年第6期1413-1421,共9页
Journal of Software
基金
Supported by the National Natural Science Foundation of China under Grant Nos.60473071,10476006(国家自然科学基金)
the National High-Tech Research and Development Plan of China under Grant No.2006AA01Z414(国家高技术研究发展计划(863))
关键词
数据流
流能量
预测查询
小波分解
频繁模式
data stream
stream energy
predictive query
wavelet analyze
frequent pattern