Many XML document clustering algorithms need to compute similarity among documents. Due to its semi-structured characteristic, exploiting the structure information for computing structural similarity is a crucial issue in XML similarity computation. Some path based approaches model the structure as path set and use the path set to compute structural similarity. One of the defects of these approaches is that they ignore the relationship between paths. In this paper, we propose the conception of F requent A ssociation T ag S equences ( FATS ). Based on this conception, we devise an algorithm named FATSMiner for mining FATS and model the structure of XML documents as FATS set, and introduce a method for computing structural similarity using FATS. Because FATS implies the ancestor-descendant and sibling relationships among elements, this approach can better represent the structure of XML documents. Our experimental results on real datasets show that this approach is more effective than the other path based approaches. © 2012 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Zhang, L., Li, Z., Chen, Q., Li, X., Li, N., & Lou, Y. (2012). Mining frequent association tag sequences for clustering XML documents. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7235 LNCS, pp. 85–96). https://doi.org/10.1007/978-3-642-29253-8_8
Mendeley helps you to discover research relevant for your work.