... also differ from Livesay and Burgess (1997) who found that mediated primes were fur- ther from their targets than unrelated controls, us- ing however a model and corpus different from the ones ... Street Sheffield S1 4DP, UK mlap@dcs.shef.ac.uk Abstract Traditional vector-based models use word co-occurrence counts from large corpora to represent lexical meaning. In this pa- per we pr...
Ngày tải lên: 17/03/2014, 06:20
Báo cáo khoa học: "Learning Lexicalized Reordering Models from Reordering Graphs" potx
... method is also general to other lexicalized reordering models. We plan to apply our method to the complex lexicalized reordering models, for example, the hierarchical reordering model (Galley and ... China {sujinsong,yliu,lvyajuan,htmi,liuqun}@ict.ac.cn Abstract Lexicalized reordering models play a crucial role in phrase-based translation systems. They are usually learned f...
Ngày tải lên: 23/03/2014, 16:20
... we used our general corpora to produce three different models: GSLCLM from the GSLC corpus, NewsLM from the newspaper corpus and DomNewsLM from the domain adapted newspaper Corpus. 3.1 Interpolating ... trigram language models we used the SRI language modelling toolkit (Stol- cke, 2002) with Good-Turing discounting. The first model was generated directly from the MP3 corpus we got...
Ngày tải lên: 22/02/2014, 02:20
Báo cáo khoa học: "Learning Condensed Feature Representations from Large Unsupervised Data Sets for Supervised Learning" docx
... models are currently the most widely-used models and are employed in many NLP systems. To simplify the explanation, we define function r(x, y), where r(x, y) returns 1 if y = ˆ y is obtained from ... approach, C is induced independently from F, and used in addition to F in supervised learning, i.e., F ∪ C. In contrast, in our method H is directly induced from F with the help of an ex...
Ngày tải lên: 07/03/2014, 22:20
Báo cáo khoa học: "Learning High-Level Planning from Text" pot
... can be made from wood. (a) Low Level Actions for: wood → pickaxe → stone step 1: move from (0,0) to (2,0) step 2: chop tree at: (2,0) step 3: get wood at: (2,0) step 4: craft plank from wood step ... relations from text and to per- form high-level planning guided by those rela- tions. We implement this idea in the reinforce- ment learning framework using feedback au- tomatically obtai...
Ngày tải lên: 16/03/2014, 19:20
Báo cáo khoa học: "Learning Translation Consensus with Structured Label Propagation" potx
... decoding for lattices and hypergraphs. Others extend consensus among translations from the same MT system to those from different MT systems. Collaborative decoding (Li et al., 2009) scores the ... , , , , , 5 where , is the propagating probability from source sentence/span to , and , is that from...
Ngày tải lên: 23/03/2014, 14:20
Báo cáo khoa học: "Learning Context-Dependent Mappings from Sentences to Logical Form" doc
... ex- 976 Example #1: (a) show me the flights from boston to philly λx.flight(x) ∧ from( x, bos) ∧ to(x, phi) (b) show me the ones that leave in the morning λx.flight(x) ∧ from( x, bos) ∧ to(x, phi) ∧ dur ing(x, ... aircraft(x) = y Example #2: (a) show me flights from milwaukee to orlando λx.flight(x) ∧ from( x, mil) ∧ to(x, orl) (b) cheapest argmin(λx.flight(x) ∧ from( x, mil) ∧ to(x, or...
Ngày tải lên: 23/03/2014, 16:21
Báo cáo khoa học: "Learning to Extract Relations from the Web using Minimal Supervision" ppt
... experimen- tal results demonstrating that our approach is able to accurately extract relations from the web by learning from such weak supervision. 2 Problem Description We address the task of learning a ... level, we manually annotated all instances from the positive test bags. The last column in Ta- bles 2 & 3 shows, between parentheses, how many instances from the positive test...
Ngày tải lên: 23/03/2014, 18:20
Báo cáo khoa học: "Learning PP attachment for filtering prosodic phrasing" potx
... Dutch PP attachment instances from a syntac- tic treebank. Collection of this data is described in Section 2. We extract lexical head features (uni- gram and bigram) from the treebank occurrences, and ... Cooccurrence strength values are typically estimated from a very large corpus. Often, these corpora are static and do not contain neologisms and names from later periods. In this pap...
Ngày tải lên: 24/03/2014, 03:20
Báo cáo khoa học: "Extracting Social Power Relationships from Natural Language" potx
... syntactic patterns of expressions on how to de- termine the polarity from its lexical components. McDonald et al studied models that classify senti- ment on multiple levels of granularity: sentence ... model personality suggests an ability to model so- cial power lects as well. Apart from text classification, work from the topic modeling community is also closely related to Socia...
Ngày tải lên: 07/03/2014, 22:20