A Flexible Structured-based Representation for XML Document Mining

Abstract

This paper reports on the INRIA group's approach to XML mining while participating in the INEX XML Mining track 2005. We use a flexible representation of XML documents that allows taking into account the structure only or both the structure and content. Our approach consists of representing XML documents by a set of their sub-paths, defined according to some criteria (length, root beginning, leaf ending). By considering those sub-paths as words, we can use standard methods for vocabulary reduction, and simple clustering methods such as K-means that scale well. We actually use an implementation of the clustering algorithm known as "dynamic clouds" that can work with distinct groups of independent variables put in separate variables. This is useful in our model since embedded sub-paths are not independent: we split potentially dependant paths into separate variables, resulting in each of them containing independant paths. Experiments with the INEX collections show good results for the structure-only collections, but our approach could not scale well for large structure-and-content collections.