Resolving XML semantic ambiguity

Abstract

XML semantic-aware processing has become a motivating and important challenge in Web data management, data processing, and information retrieval. While XML data is semi-structured, yet it remains prone to lexical ambiguity, and thus requires dedicated semantic analysis and sense disambiguation processes to assign well-defined meaning to XML elements and attributes. This becomes crucial in an array of applications ranging over semantic-aware query rewriting, semantic document clustering and classification, schema matching, as well as blog analysis and event detection in social networks and tweets. Most existing approaches in this context: i) ignore the problem of identifying ambiguous XML nodes, ii) only partially consider their structural relations/context, iii) use syntactic information in processing XML data regardless of the semantics involved, and iv) are static in adopting fixed disambiguation constraints thus limiting user involvement. In this paper, we provide a new XML Semantic Disambiguation Framework titled XSDF designed to address each of the above motivations, taking as input: an XML document and a general purpose semantic network, and then producing as output a semantically augmented XML tree made of unambiguous semantic concepts. Experiments demonstrate the effectiveness of our approach in comparison with alternative methods.