Tài liệu Báo cáo khoa học: "Generating statistical language models from interpretation grammars in dialogue systems" potx

... Generating statistical language models from interpretation grammars in dialogue systems Rebecca Jonson Dept. of Linguistics, G ¨ oteborg University and GSLT rj@ling.gu.se Abstract In this paper, ... are considering basing our re-ranking on the informa- tion held in the dialogue information state, knowl- edge of what is going on in the graphical interface and on dialog...

Ngày tải lên: 22/02/2014, 02:20

... deterministic left-corner parsing. In Proceedings of the Workshop on Incremental Parsing: Bringing Engineering and Cognition Together, pages 26–33. Liang Huang and Haitao Mi. 2010. Efficient incremental decoding for ... Yamada. 2003. Syntax-based language models for statistical machine translation. In Proceedings of the Ninth Ma- chine Translation Summit of the International Associ-...

Ngày tải lên: 20/02/2014, 04:20

... Introduction Transliteration mining is the extraction of translit- eration pairs from unlabelled data. Most transliter- ation mining systems are built using labelled train- ing data or using heuristics to ... for creating the training data described in Section 5.1.1 and build a word- aligned list and a cross-product list from the parallel corpus. We first train and test our unsupervised...

Ngày tải lên: 19/02/2014, 19:20

... available in very limited domains and building bilingual re- sources in a new domain is usually very expensive. It is an interesting question whether a model that is trained on an existing large bilingual ... Blvd, Gatineau, QC, Canada george.foster@nrc.gc.ca Abstract Statistical machine translation is often faced with the problem of combining training data from many diverse sources i...

Ngày tải lên: 19/02/2014, 19:20

... (LDC2005T10), as shown in Ta- ble 1. Using these corpora, we got three kinds of col- location models: CM-1: the training data is the additional mo- nolingual corpora; CM-2: the training data is either ... trained from the word aligned bilingual corpus. 4 Improving Phrase Table Phrase-based SMT system automatically extracts bilingual phrase pairs from the word aligned bi- lingual...

Ngày tải lên: 20/02/2014, 04:20

... Higashinaka, Rashmi Prasad, and Marilyn Walker. 2006. Learning to Generate Naturalistic Utterances Using Reviews in Spoken Dialogue Sys- tems. In COLING-ACL06, Sidney, Australia. Minqing Hu and Bing ... and Bing Liu. 2004. Mining and summa- rizing customer reviews. In Proceedings of KDD, Seattle, Washington, USA, August. Dan Klein and Christopher D. Manning. 2002. Fast exact inference...

Ngày tải lên: 20/02/2014, 04:20

... various kinds of logic models and probabilistic models (e.g., [14, 3, 15, 22]). 4. After applying the Bayes’ formula and dropping a document-independent constant (since we are only interested in ranking ... document language model, we derived a general retrieval formula where the smoothing of the document language model can be interpreted in terms of several heuristics used intrad...

Ngày tải lên: 20/02/2014, 09:20

... researchers take responsibility for maintaining their own web pages and, in addition, researchers are organised into research groups that also maintain a web page. In this framework, information easily gets ... create and maintain by hand. We are exploring the idea of automatically generated and updated web pages that accurately reflect the re- search interests being pursued within a resea...

Ngày tải lên: 20/02/2014, 09:20

... Translation Setup Starting from the collection of parallel training sen- tences, we train word alignment models in two trans- lation directions, from English to Iraqi Arabic and from Iraqi Arabic ... 1993). Following the path, we shall put more constraints on word alignment models and investigate ways of implementing them in a statistical framework. We have seen examples showi...

Ngày tải lên: 20/02/2014, 12:20

... shown in Table 5. 6.2 Training and test perplexities In order to compute the training and test perplex- ities, we split the whole aligned training corpus in two parts as shown in Table 6. The training and ... computed automatically using another statistical train- ing procedure (Och, 1999) which often pro- duces word classes including words with the same semantic meaning in the sa...

Ngày tải lên: 20/02/2014, 18:20