Semano: Semantic Annotation Framework for Natural Language Resources

In this paper, we present Semano - a generic framework for annotating natural language texts with entities of OWL 2 DL ontologies. Semano generalizes the mechanism of JAPE transducers that has been introduced within the General Architecture for Text Engineering (GATE) to enable modular development of annotation rule bases. The core of the Semano rule base model are rule templates called japelates and their instantiations. While Semano is generic and does not make assumptions about the document characteristics used within japelates, it provides several generic japelates that can serve as a starting point. Also, Semano provides a tool that can generate an initial rule base from an ontology. The generated rule base can be easily extended to meet the requirements of the application in question. In addition to its Java API, Semano includes two GUI components - a rule base editor and an annotation viewer. In combination with the default japelates and the rule generator, these GUI components can be used by domain experts that are not familiar with the technical details of the framework to set up a domain-specific annotator. In this paper, we introduce the rule base model of Semano, provide examples of adapting the rule base to meet particular application requirements and report our experience with applying Semano within the domain of nano technology.