To improve similarity search efficiency for multivariate time series datasets, distance-based index structure(Dbis) for similarity search is introduced. The dimension of MTS database is reduced by Principal Component Analysis(PCA)method, and the principal component of MTS are clustered, and the MTS items are mapped into one dimensional space based on clustering centre of each partition, on B+-tree indexing configuration, k MTS items are found out as most similar MTS sequences for given MTS sequence. Experimental results show that the proposed algorithm detects similar MTS more accurately and efficiently.%为提高多元时间序列相似查询执行效率,采用了基于距离索引结构的相似查询算法。利用主成分分析方法对多元时间序列数据降维并在此基础上进行聚类,以聚类质心为参考点,将各类变换到一维空间,利用B+-tree结构进行索引查询,找到与查询序列最相似的k个MTS序列。实验表明查询效率和准确性都有比较大的提高。
展开▼