... non-linguistic sources, such as pointing actions, were taken into account. The discourseentities are used in intra- and extra-sentential pronoun resolution in BBN Janus. 1 Introduction Discourse ... for generating DEs for dependent quantifiers stemming from indefinite and definite NPs which over- comes some difficulties in capturing dependencies be- tween discourse entities. In our multi-modal ... Joshi, Aravind K., Weinstein, Scott. (1983). Providing a Unified Account of Definite Noun Phrases in Discourse, Proceedings of the 21st Annual Meeting of the ACL. Cambridge, MA: ACL. Hinrichs,...
... discourse purpose/intention. Satisfying the main discourse purpose is achieved by satisfying several smaller purposes/intentions organized in a hierar-chical structure. As a result, the discourse ... result, the instruction did not match the initial essay quality. Nonetheless, in the open-question interviews, many users indicated using the NM as a reference while updating their essay. In addition ... analy-sis of discourse segments in direction-giving mono-logues. In Proc. of ACL. E. Hovy. 1993. Automated discourse generation using discourse structure relations. Articial Intelligence,...
... (such as acknowledging utterances by others and not inter- rupting) as well as some extra goals coming from the domain setting to maintain a shared view of the world and the domain plans which ... Domain Plan Negotiation 6. High-level Discourse Goals 4 architecture can handle varying degrees of initiative, while remaining responsive. The default behavior is to allow the user to maintain ... role in accounting for the interactions in dialog. Obligations do not replace the plan-based model, but augment it. The result- ing model more readily accounts for discourse behavior in adversarial...
... status in dialogue. Availablefrom http://www.stanford.edu/class/cs224u/guidelines-infostatus.pdf.Malvina Nissim. 2006. Learning information status of discourse entities. In Proceedings of ... the IS of discourse entities. We examine a related but more challengingtask, fine-grained IS determination, whichinvolves classifying a discourse entity asone of 16 IS subtypes. We investigate ... training/test instance fromeach hand-annotated NP in the training/test set.Each instance is represented using five types offeatures, as described below.Unigrams (119704). We create one binary...
... right sibling, Right Sib-ling Contains a VP and Right Sibling Containsa Trace.3 Discourse vs. non -discourse usageOf the 100 connectives annotated in the PDTB,only 11 appear as a discourse ... improvement over those ob-tained by Marcu (2000) in his corpus-based ap-proach which achieves an f-score of 84.9%3foridentifying discourse connectives in text. Whilebearing in mind that the evaluations ... high baseline, with an f-score of75.33% and an accuracy of 85.86%. Interest-ingly, using only the syntactic features, ignoringthe identity of the connective, is even better, re-sulting in an...
... substancesand findings have special affixes, which are verydistinguishable from ordinary words.Lexical Features: Every token in the trainingdata was used as a feature. Alphabetic words in the training data ... acronyms in the notes. This also suggest that this kind of clin-ical notes are very noisy, and require a consider-23able amount of effort in pre-processing. Allow-ing partial matching increased ... an indication that suf-ficient training data is a crucial factor in achievinghigh performance. SUBSTANCE, PROCEDURE andFINDING are the best three categories due to theirhigh frequency in the...
... and semantic role chunking using support vector machines. In HLT-03. R. Higashinaka, N. Miyazaki, M. Nakano, and K. Ai-kawa. 2004. Evaluating Discourse Understanding in Spoken Dialogue Systems. ... training corpus. From this figure, we can find that more training sentences for the se-mantic dependency graph with discourse analysis are needed than that without discourse. This im-plies discourse ... investigated. Three main services: registration information service, clinic information service, and FAQ information service are used. This system mainly provides the function of on-line registration....
... Artif. Intell., 165(1):91–134.Tim Finin, Will Murnane, Anand Karandikar, NicholasKeller, Justin Martineau, and Mark Dredze. 2010.Annotating named entitiesin twitter data with crowd-sourcing. In ... sourcedomain training set. Wu et al. (2009) propose an-other bootstrapping algorithm that selects bridginginstances from an unlabeled target domain, whichare informative about the target domain and ... Zhai. 2007. Instance weight-ing for domain adaptation in nlp. In ACL, pages 264–271.Dan Klein and Christopher D. Manning. 2003. Accurateunlexicalized parsing. In ACL, pages 423–430.Vijay...
... use cen-tering theory (Kameyama, 1986) to determine howeasily a noun phrase can be referred to in the follow-ing context.2.2 Centering TheoryCentering theory is an empirical sorting rule used ... Meeting of Association ofComputational Linguistics, pages 423–429.B. J. Grosz, A. K. Joshi, and S. Weistein. 1983. Providing aunified account of definite nounphrases in discourse. AnnualMeeting ... Association of Com-putational Linguistics, pages 178–181.M. Kameyama. 1986. A property-sharing constraint in center-ing. Annual Meeting of Association of Computational Lin-guistics, pages 200–206.T....
... principles associated with routines can determine the encoding of these routine functions in sequences of utterances. According to the model we are developing, a sequence of routine continuations ... as 243 interaction conforms to a discourse routine, which not only allows more refined analyses of routine behavior, but also permits fine-grained comparison of discourses obtained under ... comparing discourse level properties, for correlating discourse features with other textual features, and for analyses of discourse management strategies. Introduction Increasingly, research in...
... pushing causing the falling, given the discourse context in which the pushing and falling are described in (2), Narration is inferred after all, and so the falling precedes the push. In ... switch- ing (or, indeed, amelioration). In one sense this isn't surprising. Causal knowledge and MCT were in conflict in (2), and since both laws relate to the domain, but in incommensurable ... change in the discourse 'pattern' established in the preceding context. Patterns in DICE Can we use presentational constraints without ac- cidentally blocking discourse popping and inter-...
... remainder is used in training.4 ModelsOur starting point for statistical NER is a feature-based linear model over sequences, trained usingthe structured perceptron (Collins, 2002).8 In addition ... new domain, sacri-ficing performance in the original domain.The recall-oriented bias is not merely encour-aging the learner to identify entities already seen in training. As recall increases, ... adapt our model with self-trainingon unlabeled target-domain data; enforc-ing the same recall-oriented bias in the self-training stage yields marginal gains.11 IntroductionThis paper considers...
... mentioned in H can only be covered byT by resolving a bridging reference with China in T. To connect the bridging referents, a newtree component representing the bridging relationis inserted into ... to cover AS-28 mini submarine in H from the coreferring it in T , mini submarine in Tand AS-28 vehicle in T. A substitution of it byeither coreference does not suffice, since none ofthe ... role of discourse. Since in practical settings, discourse plays an important role, our goal was to developan agenda for improving the handling of discourse references in entailment-based inference.Our...
... Seg-menting and Labeling Sequence Data. In Interna-tional Conference on Machine Learning 2001, pages282–289.M. Lapata and A. Lascarides. 2004. Inferringsentence-internal temporal relations. In ... to-to the -in and-but but-the to-it* and-and the-the in- ina-but he-but said -in to-the of-and a-ofsaid-but they-but of -in in-and in- of s-andTable 1: Word pairs with highest information gain.Also ... removingword pairs that appear less than 5 times, the re-maining features were ranked by information gainusing the MALLET toolkit1.Table 1 lists the word pairs with highest infor-mation gain...
... is effective in reducing the cost of creating the discourse un-derstanding component in that no hand-crafted rulesare necessary. For statistical discourse understand-ing, an initial system, ... collected in acousti-cally insulated booths. The task domain was meet-ing room reservation. Subjects were instructed toreserve a meeting room on a certain date from a cer-tain time to a certain ... discourse understand-ing component. Since we are resolving the ambi-guity of speech understanding from the discourse point of view and not within the speech understand-ing candidates, we assume...