Multidimensional markup and heterogeneous linguistic resources

Abstract

The paper discusses two topics: firstly an approach of using multiple layers of annotation is sketched out. Regarding the XML representation this approach is similar to standoff annotation. A second topic is the use of heterogeneous linguistic resources (e.g., XML annotated documents, taggers, lexical nets) as a source for semiautomatic multi-dimensional markup to resolve typical linguistic issues, dealing with anaphora resolution as a case study. 1

Web-based annotation of anaphoric relations and lexical chains

“…Furthermore, the output of different resources often cannot be encoded in a single structure without driving into incompatibilites (i. e. XML overlap). Therefore an architecture was developed which allows for the combination of the output structures of several linguistic resources into a single XML annotated document and which is described in detail in Witt et al (2005) and Stührenberg et al (2006) .…”

Different Views on Markup

“…The discourse parser described is realised in Prolog and takes the Prolog fact base derived from the multiple annotations of one document as its input and adds the independent annotation layer of rhetorical structure as its output. Stührenberg et al (2006) apply the framework within the context of anaphora resolution. Necessary resources for the resolution process (e.g.…”

Freestyle Markup Language

“…For an FML graph an unlimited number of perspective-nodes -sharing the same document content-nodes -will directly follow the document-root-node. This important topic has been discussed for a long time [Stührenberg 2006] [ Witt 2007] and there are plenty of scenarios that would benefit from markup ability to handle independent structures. For instance, the American National Corpus [1] would not have to redundantly administrate their linguistic inventory seven times stand-off [Ide 2006].…”