A statistical interestingness measures for XML based association rules
PRICAI'10 Proceedings of the 11th Pacific Rim international conference on Trends in artificial intelligence
Hi-index | 0.00 |
The growing research interests in the XML data warehousing and XML mining areas during the last few years were determined by the wider use of the XML to represent semi-structured data and to exchange information between different types of applications. A large number of techniques have being developed, to mine interesting knowledge from XML documents, e.g. frequent patterns, association rules, clusters etc. Though, there was little or no attention accorded to the changes affecting the discovered knowledge, in cases when the initial source of data, i.e. the XML documents, was changing in time. In this paper we propose a novel technique of determining how the knowledge discovered from initial XML documents changes in time when the documents' content and/or structure fluctuates.