Abstract: A query processor called lazyDFA which is created based on the automaton technology is an effective method to solve the problem of XPath query on XML data stream.This paper proposes the optimized algorithms for lazyDFA which use the semantic and structural information of DTD in order to improve the efficiency of lazyDFA.The experimental results show that optimized algorithms can effectively accelerate lazyDFA query speed，lower memory consumption.

Key words: Document Type Definition（DTD）, lazyDFA, query optimization, XML data stream

摘要： 针对XML数据流上的XPath查询处理问题，基于自动机技术构造的lazyDFA查询处理器是一种有效的解决方法。为了提高lazyDFA的查询处理效率，提出了利用DTD的语义信息和结构信息对lazyDFA进行优化的算法。实验结果表明，优化算法可以有效的加快lazyDFA的查询速度，降低内存消耗。

关键词: 文档类型定义, lazyDFA, 查询优化, XML数据流