File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/relat/05/w05-0303_relat.xml
Size: 7,792 bytes
Last Modified: 2025-10-06 14:15:50
<?xml version="1.0" standalone="yes"?> <Paper uid="W05-0303"> <Title>Referential Annotations</Title> <Section position="6" start_page="16" end_page="17" type="relat"> <SectionTitle> 5 Related Work </SectionTitle> <Paragraph position="0"> This section discusses how the unified SYN-RA annotation scheme relates to other formats currently discussed in the literature, in particular the pie-in-the-sky scheme for semantic annotation5 and the annotation graph model of (Bird and Liberman, 2001). While these two annotation schemes are by no means the only contenders for corpus annotation standards in the literature, they are certainly among the most ambitious and promising.</Paragraph> <Paragraph position="1"> While the pie-in-the-sky scheme is clearly still under development, the following characteristics and goals can already be gleaned from its web-page and the annotation examples presented there: The annotation is feature-structure-based and incorporates various levels of linguistic annotation, in particular a PROPBANK style predicate-argument structure, dependency style syntactic information, as well as morpho-syntactic and word class information. All this information is rooted in the attributes needed for predicate-argument assignment, <sentence id=&quot;s11976&quot;> <node id=&quot;s11976n518&quot; cat=&quot;SIMPX&quot; func=&quot;--&quot; parent=&quot;0&quot;> <node id=&quot;s11976n515&quot; cat=&quot;VF&quot; func=&quot;-&quot;> <node id=&quot;s11976n513&quot; cat=&quot;NX&quot; func=&quot;OA&quot;> <node id=&quot;s11976n500&quot; cat=&quot;NX&quot; func=&quot;APP&quot;> <word id=&quot;s11976w0&quot; form=&quot;Ihre&quot; pos=&quot;PPOSAT&quot; morph=&quot;asf&quot; func=&quot;-&quot;> < anaphora> < relation type=&quot;ana&quot; antecedent=&quot;s11975n517&quot;/> < /anaphora> </word> <word id=&quot;s11976w1&quot; form=&quot;Schulkameradin&quot; pos=&quot;NN&quot; morph=&quot;asf&quot; func=&quot;HD&quot;/> </node> <node id=&quot;s11976n508&quot; cat=&quot;EN-ADD&quot; func=&quot;APP&quot;> <node id=&quot;s11976n501&quot; cat=&quot;NX&quot; func=&quot;-&quot;> <word id=&quot;s11976w2&quot; form=&quot;Cassie&quot; pos=&quot;NE&quot; morph=&quot;asf&quot; func=&quot;-&quot;/> <word id=&quot;s11976w3&quot; form=&quot;Bernall&quot; pos=&quot;NE&quot; morph=&quot;asf&quot; func=&quot;-&quot;/> </node> </node> </node> </node> <node id=&quot;s11976n509&quot; cat=&quot;LK&quot; func=&quot;-&quot;> <node id=&quot;s11976n502&quot; cat=&quot;VXFIN&quot; func=&quot;HD&quot;> <word id=&quot;s11976w4&quot; form=&quot;fragten&quot; pos=&quot;VVFIN&quot; morph=&quot;3pit&quot; func=&quot;HD&quot;/> </node> </node> <node id=&quot;s11976n510&quot; cat=&quot;MF&quot; func=&quot;-&quot;> <node id=&quot;s11976n503&quot; cat=&quot;NX&quot; func=&quot;ON&quot;> <word id=&quot;s11976w5&quot; form=&quot;sie&quot; pos=&quot;PPER&quot; morph=&quot;np*3&quot; func=&quot;HD&quot;> < anaphora> < relation type=&quot;ana&quot; antecedent=&quot;s11976w1&quot;/> < /anaphora> </word> </node> </node> <word id=&quot;s11976w6&quot; form=&quot;,&quot; pos=&quot;$,&quot; morph=&quot;--&quot; func=&quot;--&quot; parent=&quot;0&quot;/> <node id=&quot;s11976n517&quot; cat=&quot;NF&quot; func=&quot;-&quot;> <node id=&quot;s11976n516&quot; cat=&quot;SIMPX&quot; func=&quot;OS&quot;> <node id=&quot;s11976n504&quot; cat=&quot;C&quot; func=&quot;-&quot;> <word id=&quot;s11976w7&quot; form=&quot;ob&quot; pos=&quot;KOUS&quot; morph=&quot;--&quot; func=&quot;-&quot;/> </node> <node id=&quot;s11976n514&quot; cat=&quot;MF&quot; func=&quot;-&quot;> <node id=&quot;s11976n505&quot; cat=&quot;NX&quot; func=&quot;ON&quot;> <word id=&quot;s11976w8&quot; form=&quot;sie&quot; pos=&quot;PPER&quot; morph=&quot;nsf3&quot; func=&quot;HD&quot;> < anaphora> < relation type=&quot;ana&quot; antecedent=&quot;s11976n513&quot;/> < /anophora> </word> </node> <node id=&quot;s11976n511&quot; cat=&quot;PX&quot; func=&quot;OPP&quot; comment=&quot;&quot;> <word id=&quot;s11976w9&quot; form=&quot;an&quot; pos=&quot;APPR&quot; morph=&quot;a&quot; func=&quot;-&quot;/> <node id=&quot;s11976n506&quot; cat=&quot;NX&quot; func=&quot;HD&quot;> <word id=&quot;s11976w10&quot; form=&quot;Gott&quot; pos=&quot;NE&quot; morph=&quot;asm&quot; func=&quot;HD&quot;/> </node> </node> </node> <node id=&quot;s11976n512&quot; cat=&quot;VC&quot; func=&quot;-&quot;> <node id=&quot;s11976n507&quot; cat=&quot;VXFIN&quot; func=&quot;HD&quot;> <word id=&quot;s11976w11&quot; form=&quot;glaube&quot; pos=&quot;VVFIN&quot; morph=&quot;3sks&quot; func=&quot;HD&quot;/> </node> </node> </node> </node> </node> <word form=&quot;.&quot; pos=&quot;$.&quot; morph=&quot;--&quot; func=&quot;--&quot; parent=&quot;0&quot;/> A graphical representation of the relation is shown in Figure 3. The antecedent &quot;s3426n507&quot; refers to the first NP, &quot;s3426n526&quot; to the second one in Figure 3.</Paragraph> <Paragraph position="2"> with syntactic and morpho-syntactic information distributed among the corresponding elements in the predicate-argument structure representation. Accordingly, semantic representations provide the organizing principle while morpho-syntactic and syntactic information play a subordinated role.</Paragraph> <Paragraph position="3"> The SYN-RA annotation scheme resembles the pie-in-the-sky scheme in that it also uses one level of representation, in this case hierarchical syntactic structure, as the organizing principle and treats referential relations, grammatical function information, and morpho-syntactic annotation as subordinated types of information. More generally, the pie-in-the-sky and the SYN-RA representations offer a particular view of the annotation, each with its own &quot;perspective&quot;: semantics-based (pie-in-the-sky) and syntax-based (SYN-RA).</Paragraph> <Paragraph position="4"> By contrast, Bird and Liberman's (2001) annotation graphs are intended as a graph-based, multi-layered annotation scheme where each level of linguistic annotation is treated equally, as an independent layer. The graph-based annotation model is powerful enough to also allow groupings of discontinuous constituents and other non-adjacent linguistic phenomena, without having to rearrange the linear order of the input. In both respects, their annotation model is maximally general.</Paragraph> </Section> class="xml-element"></Paper>