FootbOWL: Using a generic ontology of football competition for planning match summaries

Presented at: 8th Extended Semantic Web Conference (ESWC2011)

by Nadjet Bouayad-Agha, Gerard Casamayor, Leo Wanner, Sergio López Hernández

We present an approach to content selection and discourse structuring in text generation that work directly on a task-independent domain knowledge base (KB) modeled in OWL. In order to facilitate the representation of objects, events and semantic relations that are inferred from the basic facts of the domain to obtain richer and more fluent texts, we distinguish between an extended ontology and an upper layer ontology. The nodes in the KB are weighted according to learnt models of content selection, such that a subset of them can be extracted. The extraction is done using templates that also consider semantic relations between the nodes and the user profile. The discourse structuring submodule maps the semantic relations to discourse relations and forms discourse units to then arrange them into a coherent discourse graph. The approach is illustrated and evaluated on an ontology that models the First Spanish Football League.

Keywords: inference rules, natural language generation, ontology, text planning

Resource URI on the dog food server:

Explore this resource elsewhere: