摘要
为了有效进行时间序列的匹配,提出了基于交叉覆盖算法的模式匹配方法。首先对时间序列进行预处理,使它们分布在n+1维空间中某个中心在原点的球面上,再通过学习构造分类器,实现序列的准确匹配。主要讨论完全匹配查询,即把两两相似的时间序列分别归类,所以利用交叉覆盖算法准确分类的功能可以完成所探讨的问题。
Time series pattern matching is an important issue in data mining. In this paper, a new matching method based on alternative covering algorithm was proposed. After pretreatment, time series were distributed over a spherical surface with its center at orion. Then a classifier was constructed through learning and pattern matching can be well and truly finished. All-palrs query was mainly discussed in this paper, that is to say, classifying similar time series into the same category was the primary task. Pattern matching discussed here can be effectively fulfilled by the matching method based on alternative covering algorithm, which has the function of classifying samples accurately.
出处
《计算机应用》
CSCD
北大核心
2007年第2期425-427,共3页
journal of Computer Applications
基金
安徽省自然科学基金资助项目(050460402)
安徽省教育厅科研项目(2005kj056)
安徽省高等学校青年教师科研资助计划(2005jq1035)
关键词
交叉覆盖算法
模式匹配
分类
完全匹配查询
alternative covering algorithm
pattern matching
classifying
all-pairs query
