Báo cáo khoa học: "Labeling Documents with Timestamps: Learning from their Time Expressions" pot

... 8-14 July 2012. c 2012 Association for Computational Linguistics Labeling Documents with Timestamps: Learning from their Time Expressions Nathanael Chambers Department of Computer Science United ... Before Timestamp: M < T 2. After Timestamp: M > T 3. Same as Timestamp: M == T We also experiment with 7 fine-grained relations: 1. One year Before Timestamp: M == T − 1 2. T...

Ngày tải lên: 07/03/2014, 18:20

... drawn from BC. We then investigate the utility of adding additional in-domain training data from WSJ. In the baseline approach, the additional WSJ examples are randomly selected. With active learning ... with a naive Bayes classifier, the sense s assigned to an example with features f 1 , . . . , f n is chosen so as to maximize: p(s) n  j=1 p(f j |s) In our domain adaptation study...

Ngày tải lên: 08/03/2014, 02:21

... unification-based framework with large-coverage gram- mars and how from their usage lexi- cal entries are extracted. To keep the time and space usage during parsing within bounds, information from exter- nal ... Starting from this "standard" tech- nique, we extend it and integrate PoS and mor- phological information, originating from external resources. We will first d...

Ngày tải lên: 17/03/2014, 22:20

... description of the intended referent(s), the target set. Its elements are to be distinguished from potential distractors (McDonald 1981), the contrast set, which entails all elements of the context ... and fast strategy of incre- mentally selecting ambiguity-reducing attributes from a domain-dependent preference list. With the extension of the algorithm's scope to the identific...

Ngày tải lên: 22/02/2014, 02:20

... specialist medical terms, the latter has well over 100,000 entries, gathered from machine-readable sources, however, their grammar formalism and the limited grammatical information supplied by ... ul[Tl;X9]tocauseto ~sten with RIVETsI: 28289801<RO154300<rlvet 28289902<02< < 28290005<v< 28290107<0100<TI;X9<NAZV< H XS 28290208<to cause to faste...

Ngày tải lên: 22/02/2014, 09:20

... buffer solutions: 100 mm Hepes from pH 6.8 to pH 8.2, 100 mm EPPS from pH 7.6 to pH 8.6, and 100 mm TAPS (N- [tris(hydroxymethyl)methyl]-3-aminopropanesulfonic acid) from pH 7.7 to pH 9.1. The E. ... was withdrawn, and the residual activity was measured using the direct en- zymatic assay, as described above. At the same time, the remaining mixture was diluted to a final volume of 100 lL...

Ngày tải lên: 07/03/2014, 11:20

... consistent with our intuition: recall can be improved with a larger training set, but preci- sion can be hampered when learning from nois- ily labeled data. Overall, these results suggest that learning ... we augment P with all the unlabeled documents con- taining at least three words from P W . The rea- son we impose the “at least three” requirement is precision: we want to e...

Ngày tải lên: 08/03/2014, 00:20

... synsets gathered for a FrameNet more of the verbs from the SemFrame frame are frame name intersects with the set of synsets semantically related to verbs from the FrameNet gathered for a SemFrame frame ... Compared with FrameNet 7 Conclusions and Future Work We have demonstrated that sets of verbs evoking a common semantic frame can be induced from existing lexical tools. In a head-...

Ngày tải lên: 17/03/2014, 06:20

... were marked with special tags to distinguish them from regular links, we can selectively remove them and check the impact on MT performance. Figure 1 shows an example sentence with hu- man ... similar corpora. Without it this work would not be possible. 5 Conclusions Our experiments showed that even with human alignment, further improvement in MT quality will be small with...

Ngày tải lên: 23/03/2014, 16:20

... needs grow with our understanding of language. A number of current approaches to satisfying the lexical re- quirements for artificial devices do not involve un- supervised learning from examples. ... and the growth of linguistics, and even with the lag their maintenance will be expensive. By contrast, dictionaries constructed by automated learners from real sentences will no...

Ngày tải lên: 24/03/2014, 05:21