【摘要】 极大频繁子树挖掘在Web挖掘、HTML/XML文档分析、生物医学信息处理等领域有着重要的应用,可用于解决这些领域的自同构问题。本文提出了一种极大频繁子树挖掘算法(MFTM)。MFTM基于最右路径扩展技术,在搜索过程中,采用覆盖定理进行裁剪,压缩搜索空间,从而极大地加快了算法的收敛速度。性能实验表明,极大频繁挖掘等算法是有效和可伸缩的。更多还原

【Abstract】 A novel algorithm called Maximum Frequent Tree Mining(MFTM)is presented to discover maximum frequent sub-trees from forest.MFTM uses the right-most path expansion technique.The Overlay Theorem is proposed to reduce the search space and accelerate the convergence speed.We conduct detailed experiments to test the performance and scalability of the methods.The experiments demonstrate that MFTM is effective and scalable.MFTM can be applied to solve the isomorphic problems in the domains such as Web mining,HTML/XML data analysis,bioinformatics,and so on.更多还原