作者: A. Meisels , M. Orlov , T. Maor
DOI: 10.1109/WISEW.2002.1177861
关键词: Probability distribution 、 Xml data 、 Association rule learning 、 Data content 、 XML 、 XML validation 、 Computer science 、 Inference 、 Data mining
摘要: Knowledge inference from semi-structured data can utilize frequent sub structures, in addition to frequency of items.In fact, the working assumption present study is that sub-trees XML represent sets tags (objects) aremeaningfully associated. A method for extracting presented. It uses thresholds on frequenciesof paths and multiplicity data. The are extracted counted a procedure has O(n2) complexity. content sub-trees, form attribute values, cast tabular form. This enables search forassociations Thus, complete structure extract association rules semi-structureddata. large industrial example used demonstrate operation proposed method.