... Conundrumsinnoun phrase coreference resolution: Making sense of the state-of-the-art. In Proceedings of the Joint Conference ofthe 47th Annual Meeting of the ACL and the 4th In- ternational Joint ... for mining easypositive instances, in an attempt to avoid the inclu-sion of hard training instances that may complicatethe acquisition of an accurate coreference model.3.1.2 Training a Coreference ... Aravind K. Joshi, and Scott Wein-stein. 1983. Providing a unified account of definite noun phrases in discourse. In Proceedings of the21st Annual Meeting of the Association for Compu-tational Linguistics,...
... areunambiguously conjoined heads. We leave this to future work.2We did not include coordinate head nouns from base noun phrases (NPB) (i.e. noun phrases that do not dominate other noun phrases) because ... to measurewhich were the head nouns being conjoined in cer-tain types of coordinate noun phrase. In this paper we look at different measures ofword similarity in order to discover whether they ... pairs which make up the head wordsof two conjuncts in coordinate noun phrases (againnot including coordinate NPBs). We found 601 suchcoordinate noun pairs. The second data set consistsof 601...
... Training set size(words) Precision 25k 88.7% 200k 91.8% Number of Appearances in Training Set • • 4- - • Machine Students Figure 3: Test Set Recall vs. Frequency of Appearances in Training ... to a machine at learning linguistic information from an annotated cor- pus, and more importantly to begin to explore the differences in learning behavior between hu- man and machine. Although ... extraction of terminological noun phrases. In Proceedings of the 30th Annual Meeting of the Association of Computational Linguistics, pages 977-981. Association of Computational Linguistics....
... set with increasing amounts of training data. In all cases for the table,we are evaluating on the same thirty document test set which is included in our training set, since our system in unsupervised. ... assignment” in that reference).4 Inference DetailsUp until now, we’ve discussed Gibbs sampling, butwe are not interested in sampling from the poste-rior P (Z|X), but in finding its mode. Instead ... observing the within-document gainsfrom sharing shown in section 3, we can manuallyinspect the most frequently occurring entities in ourcorpora. Table 3 shows some of the most frequentlyoccurring...
... parsing procedure. Complex noun sequences. Semantic patterns in complex noun phrases fall into two types: part names and other noun phrases. Names for pieces of equipment often con- tain ... bracket complex noun phrases. This is the subject of the next section. HI SEMANTIC PATTERNS IN COMPLEX NOUN SEQUENCES Noun phrases in the equipment failure messages typ- ically include numerous ... tences in the medical domain were analyzed and 123 in the Navy domain. The statistics are presented in Tables 1 and 2. In particular, there were significantly more noun modifiers of nouns...
... compatiblewith the information principle. Moreover, they think the use of pronoun-headed noun phrases and noun- headed phrases are different in the type of text. They think in conversation pronouns tend ... New information is generally introduced later in theclause , using noun- headed phrases( usually indefinite in form). Due to my view, thefrequency of noun- headed phrases should be higher in object ... noun phrase. A nounin the genitive may function asdeterminer in the noun phrase. Since nounphrase has different functions they areabundant in any text. If we take away the noun phrases from...
... Meaningful Subcompounds Figure 1: General Processing for Phrase Generation After preprocessing, the system works in two stages parsing and generation. In the parsing stage, each simplex noun ... and efficient noun- phrase analy- sis techniques to enhance phrase- based IR. In par- ticular, we explored an extension of the ~phrase- based indexing in the CLARIT TM system ° using a hybrid ... meaning- ful (continuous or discontinuous) subcompounds from complex noun phrases exploiting both corpus- statistics and linguistic heuristics. Using such sub- compounds rather than whole noun...
... that of linguistic head nouns and modi-fiers for noun phrases. In many cases, the IHs of noun phrase queries are exactly the head nouns in the linguistic sense. Exceptions mostly occur in queries ... evolve, it is in- creasingly believed that search will be shiftingaway from “ten blue links” toward understandingintent and serving objects. This trend has beenlargely driven by an increasing amount ... accurate in identifying head nouns. In identifying IHs innounphrase queries, how-ever, direct syntactic analysis is unlikely to be aseffective. This is because syntactic structures are in general...
... set of coreference clusters(chains).25. Scoring. Finally, during testing Reconcileruns scoring algorithms that compare thechains produced by the system to the gold-standard chains in the ... clustering withsupport vector machines. In Proceedings of the Twenty-second International Conference on Machine Learning(ICML 2005).A. Haghighi and D. Klein. 2007. Unsupervised Coreference Resolution ... Information into Information Extraction Sys-tems by Gibbs Sampling. In Proceedings of the 21st In- ternational Conference on Computational Linguistics and44th Annual Meeting of the ACL.T. Finley...
... was, once again, to measurethe relative informativeness of the nouns in NN com-pounds. Smaller synset counts indicate more spe-cific meaning of the noun, and thus perhaps moreinformation content. ... analyzed in Bell and Plag (2010) as potential predictors ofprominence assignment in compound nouns. Likeus, they too found such features to be predictiveof prominence assignment and that combining ... variability in prominence assignment.Our results also highlight the difference betweenbeing able to use lexical information in prominenceprediction of NN compounds, or not. Using lexicalfeatures,...