Graphical Annotation for Syntax-Semantics Mapping

Graphical Annotation for Syntax-Semantics Mapping Kôiti Hasida Social ICT Research Center, Graduate School of Information Science and Technology, The University of Tokyo 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan. E-mail: hasida.koiti@i.u-tokyo.ac.jp Abstract A potential work item (PWI) for ISO standard (MAP) about linguistic annotation concerning syntax-semantics mapping is discussed. MAP is a framework for graphical linguistic annotation to specify a mapping (set of combinations) between possible syntactic and semantic structures of the annotated linguistic data. Just like a UML diagram, a MAP diagram is formal, in the sense that it accurately specifies such a mapping. MAP provides a diagrammatic sort of concrete syntax for linguistic annotation far easier to understand than textual concrete syntax such as in XML, so that it could better facilitate collaborations among people involved in research, standardization, and practical use of linguistic data. MAP deals with syntactic structures including dependencies, coordinations, ellipses, transsentential constructions, and so on. Semantic structures treated by MAP are argument structures, scopes, coreferences, anaphora, discourse relations, dialogue acts, and so forth. In order to simplify explicit annotations, MAP allows partial descriptions, and assumes a few general rules on correspondence between syntactic and semantic compositions. Keywords: diagrammatic annotation, syntax-semantics mapping, standardization 1. Introduction A potential work item (PWI) for ISO standard (let us call it ‘MAP’ for convenience in the rest of the paper) of linguistic annotation concerning syntax-semantics mapping is introduced, which is an extension of SemAF-DS (ISO, 2013), which in turn is based on Linguistic DS (Description Scheme) in ISO/IEC (2004). Importing more from Linguistic DS, MAP extends some standards devised by ISO/TC37/SC4, including LAF (Linguistic Annotation Framework; ISO 2010), SynAF (Syntactic Annotation Framework; ISO 2012a), and SemAF (Semantic Annotation Framework; ISO 2012b, 2012c, 2013), while incorporating insights from relevant literature (Asher & Lascarides 2003; Carlson, et al., 2003; Haji, et al. 2006; Mann & Thompson 1988; Palmer, et al. 2005; Prasad, et al. 2008; PTB). MAP defines how to diagrammatically annotate linguistic data to specify a mapping between its possible syntactic and semantic structures. The syntactic structures may be dependencies, coordinations, ellipses, and so forth, encompassing both intrasentential transsentential constructions, and so forth. The semantic structures consist of argument structures, scopes (of quantifications, negations, modal operators, etc.), coreferences, anaphora, and so on. A major purpose of MAP is to facilitate collaborations among people involved in research, standardization, and practical use of linguistic annotation. For that sake, MAP provides a diagrammatic sort of concrete syntax (ISO, 2012b, 2012c) for linguistic annotations far easier to understand than traditional textual concrete syntax such as in XML. Besides being diagrammatic and intuitive, MAP is formal in the same sense that UML is formal. Namely, a MAP diagram accurately specifies a mapping between syntactic structures and semantic structures of the annotated linguistic data in question. The rest of the paper is organized as follows. Section 2 introduces MAP diagrams to represent annotated linguistic data. Section 3 and 4 discuss further details of annotations concerning local and nonlocal compositions, respectively. Section 5 concludes the paper. 2. Annotated …

Content truncated to 500 words