Báo cáo khoa học: "Unsupervised Morphology Rivals Supervised Morphology for Arabic MT" pot

... 0. For both DLM_1 and DLM_2, models with the parsing weight achieve better performance; we 7 For a detailed description of the baseline trigram model, see Gao et al. (2002a). therefore ... the CER of the hy- potheses with the minimum number of errors) is presented in Table 1, indicating the upper bound on performance. We also note in Table 1 that the per- formance of the conver...

Ngày tải lên: 17/03/2014, 06:20

... BLEU perfor- mance improves from for the initial model to about for the final model (we do not include the test data block sequences in the training). Ta- ble 3 shows a typical learning curve for ... dis- criminative training and decoding performance on the training data is improved after each iteration (along with the test data decoding performance). A theoretical justification for the...

Ngày tải lên: 31/03/2014, 01:20

... Nonparametric Bayesian Models for Unsupervised Even t Corefer- ence Resolution. In Advances in Neural Information Processing Systems 23 (NIPS). Cosmin Adrian Bejan. 2007. Deriving Ch ronologi- cal Information from ... Association for Computational Linguistics Unsupervised Event Coreference Resolution with Rich Linguistic Features Cosmin Adrian Bejan Institute for Creative Technologies U...

Ngày tải lên: 07/03/2014, 22:20

... used heavily for model design and hyperparameter selec- tion. In section 5, we present final results for new test data from MUC-6 on which no tuning or devel- opment was performed. This test data will form ... values for these coefficients were not changed for the experiments on the test sets. 5.1 MUC-6 Our main evaluation is on the standard MUC-6 for- mal test set. 10 The standard ex...

Ngày tải lên: 08/03/2014, 02:21

... ACL and AFNLP Unsupervised Learning of Narrative Schemas and their Participants Nathanael Chambers and Dan Jurafsky Stanford University, Stanford, CA 94305 {natec,jurafsky}@stanford.edu Abstract We ... and have proved amenable to supervised learning from corpora like PropBank (Palmer et al., 2005) and Framenet (Baker et al., 1998). Un- fortunately, creating these supervised corpora is an...

Ngày tải lên: 23/03/2014, 16:21

... in per- formance for all models when using gold standard parses. On the gold/auto dataset, F1 increases by 2.7% for the split-merge algorithm, 2.7% for the logistic classifier, and 5.5% for the ... Given the data requirements for super- vised systems and the current paucity of such data, unsupervised methods offer a promising alternative. They require no human effort for training t...

Ngày tải lên: 30/03/2014, 21:20

... perfor- mance with TFM, using the best parameter com- binations for each corpus making relatively few modifications: the left side of Figure 2 shows a rapid performance increase, particularly for ... by this pseudocode: VOCABULARY ADDITIONS: for each class C: for each year y: PreModList(C,y,L) = OddsRatio(C,y,L) ModifyList(y) = DecisionRule(PreModList(C,y,L)) for each term k in Modif...

Ngày tải lên: 08/03/2014, 04:22

... for ranking parses in syntactic analysis. For example, Mager- man (1995), Collins (1996), and Charniak (1997) proposed statistical parsing models which incor- porated lexical/semantic information. ... sets of active fea- tures for verbs Vl, , Vm, respectively: 1. Initially, for each verb vi, set ~ as the most gen- eral one `sis of the sequence `sio, , `sire. 2. For each v...

Ngày tải lên: 08/03/2014, 06:20

... University of California, Los Angeles, California. Wu, Zhibiao and Martha Palmer. 1994. Verb se- mantics and lexical selection. In Proceedings of the 32nd Annual Meeting of the Associa- tion for Computational ... Association for Computational Linguistics, pages 1562–1572, Uppsala, Sweden, 11-16 July 2010. c 2010 Association for Computational Linguistics Plot Induction and Evolutionary...

Ngày tải lên: 17/03/2014, 00:20

... DECODE PATH(x, n, start, go) for each label y 1 node[0][y 1 ].cost ← s(y 1 , 0) node[0][y 1 ].end ← start; for j ← 1 to n − 1 for each label y j+1 bestc ← −∞; end ← null; for each label y j cost ... mentions of LEFT/RIGHT, • For each occurrence of LEFT/RIGHT, look for an ordinal number, LAST, or END (= end of the hallway) nearby, • Also, for each LEFT/RIGHT, look for a men-...

Ngày tải lên: 17/03/2014, 04:20