An ontology-driven process for unification of XML instances

XML documents have been widely used for data representation and data interchange, specially on the Web. However, the great number of XML data sources on the Web drawn from the same domain with heterogeneous representations becomes very complex the semantic integration of XML instances for querying purposes, for example. This paper presents an automatic process for the unification of XML documents representing data instances on a same domain. It receives a set of XML instances as input and generates a unique representation for them. Different from related work, the process considers a domain ontology as a basis for nomenclature and structure uniformization of the resulting XML instance. The process core is a set of unification operators that are invoked to provide the necessary adjustments in the instances in order to fit well with the domain ontology schema.