Seminar — Prof. Maria Teresa Pazienza

The multilingual aspects which characterize the (Semantic) Web and the demand for more easy-to-share forms of knowledge representation, (equally accessible by humans and machines) depict a scenario where formal semantics must coexist side-by-side with natural language, both contributing to the sharing of the content they describe. The role of different cultures and languages is fundamental in a real World aWare Web and, though English is widely accepted as a “lingua franca” all over the world, much effort could be spent to preserve further idioms (they express different cultures). As a consequence, multilinguality is perceived as one of the most relevant challenges for the Semantic Web.

These premises suggest that semantic web ontologies, delegated to express machine-readable information on the Web, should be enriched to both cover formally expressed conceptual knowledge and expose its content in a linguistically motivated fashion.
Even more could be done: revisiting ontology development process under this perspective, would in fact guarantee this scenario to become a suitable framework upon which even machine oriented tasks, like mediation and discovery, would benefit of this improved expressivity.

With such an objective we have defined OntoLing, a Protégé plug-in offering a modular and scalable framework for supporting manual annotation of ontological data with information from different, heterogeneous linguistic resources. An improved version of OntoLing prompts the user with automatic suggestions for enriching ontologies with linguistic content. It will be described how and why different kinds of linguistic enrichment processes could be performed; both algorithmic aspects and profiling of user interaction in the context of OntoLing framework will be analyzed.

The expression “Linguistic Enrichment of Ontologies” has been introduced to identify a series of different processes sharing the common objective of improving the linguistic expressivity of an ontology, through the exploitation of existing Linguistic Resources (LRs, from now on). The nature of this “linguistic expressivity” strongly depends on the LRs used for linguistic enrichment and on the specific goals the enrichment process will achieve. A few different enrichment tasks, together with possible scenarios in which these tasks may be applied, will be described.