Báo cáo khoa học: "Approximating Context-Free Grammars with a Finite-State Calculus" docx

... For appropriateness labels we turn to human annotators. Two annotators were in- structed to label a candidate as appropriate if that candidate was believable as an answer, even if that candidate ... in the face of a large imbalance in positive and negative examples. In answer typing we typically have far more inap- propriate candidates than appropriate candidates, and this is especially t...

Ngày tải lên: 08/03/2014, 21:20

... Perl. 23 ate He He:1:2-ate:2:0 ate meat ate:1:0-meat:2:1 ate with ate:1:0 -with: 2:1 meat the the:1:2-meat:2:0 with fork with: 1:0-fork:2:1 fork a a:1:2-fork:2:0 Figure 4: Examples of bigram-subtrees ate meat with ate:1:0-meat:2:1 -with: 3:1 ate with ... that our system performed better than Huang2009. Com- pared with the approach of Huang et al. (2009), our approach used additi...

Ngày tải lên: 16/03/2014, 23:20

... structures. The main achievement of DART was the integration of tradi- tional Computer Assisted Learning (CAL~ facilities - such as the ones available in the PLATO system - in an Artificial Intelligence ... education- al settings and in projects of natural language translation. Practically, our program is one of the few Intelligent Systems available in the field of Fo- reign Language...

Ngày tải lên: 18/03/2014, 02:20

... syntactic and semantic ambiguity as part of the same parse chart. These parse charts can be seen as regular tree grammars that accept the lan- guage of parse trees, and conceivably an RTG that describes ... po- sitions in a string. Tree automata are related to tree transducers as used e.g. in statistical machine trans- lation (Knight and Graehl, 2005) exactly like finite- state string auto...

Ngày tải lên: 23/03/2014, 17:20

... Dialogue Act Tagging with Transformation-Based Learning Ken Samuel and Sandra Carberry and K. Vijay-Shanker Department of Computer and Information Sciences University of Delaware Newark, ... template instan- tiations are randomly selected, where R is a pa- rameter that is set in advance. As long as R is large enough, there doesn't appear to be any significant degradation ......

Ngày tải lên: 23/03/2014, 19:20

... under consideration. For in- stance, the generalized initial and final contexts of &apos ;a patch' are Initial context Final context p &apos ;a patch' SA SB p SA SB p FOCUS FA FB 0 0 ... MA Ms 0/0 MA MB 1/1 where FA, MA, etc. are variables. Among other things this rule implies that &apos ;a patch' may be used whatever the current focus values for SA and SB, but that...

Ngày tải lên: 24/03/2014, 03:20

... the performance of automatic and human classification in three different studies. We found that automatic classification can be as good as human classification; however, the accura- cy is still ... two-way classifi- cation task of labeling sarcastic and non-sarcastic tweets. We asked three other judges to classify 10% of our S-NS dataset (i.e, 180 tweets) into sar- castic and non-sarcast...

Ngày tải lên: 20/02/2014, 05:20

... Spanish and Italian (sec- tion 3). Then we present in sections 4 and 5 a com- parative evaluation experiment proving that a hy- brid approach leads to more accurate results than a classical approach ... speakers. 1.b) play role: What role can Canada’s immigration program play in help- ing developing nations ? 1.c) make mistake: We could look back and probably see a lot of mistakes tha...

Ngày tải lên: 17/03/2014, 04:20

... grammar. An exam- ples is: S ~ al S S~al A1 S-+anS S-+anAn A~ -+ a~ X A2 + al A2 An -~ al An X-+e A1 -+ a2 Az A1 ~ an A1 A2 -+ a2 X A2 ~ an A2 An -+ a2 A, ~ An ~ an X Here the grammar ... be applied to these formalisms too. 2 Finite-state calculus A &apos ;finite-state calculus' or 'finite automata toolkit' is a set of programs for manipulating...

Ngày tải lên: 31/03/2014, 21:20

... set of fragments where each member is guaranteed to appear at least twice in the training set. In order to avoid unmanageably large grammars only maximal fragments are retained in each pair- wise ... gram- mar induction was tackled with random selection of fragments or weak constraints that led to massive grammars. A more principled technique is to use a sparse nonparametric prior, a...

Ngày tải lên: 07/03/2014, 18:20