Beijing, August 2010 An Empirical Study on Web Mining of Parallel Data Gumwon Hong 1 , Chi-Ho Li 2 , Ming Zhou 2 and Hae-Chang...

webpages, we extend our harvesting technique to create parallel data for resource limited languages in an unsupervised manner.../09/06/september-2011-web-server-survey.html 201 language pairs. As a result, web harvesting of parallel text has been...

corpora for certain language pairs, such as Spanish or French, are widely available, but for many language pairs, such as Bengali...

using English-German, English-Greek and English-Latvian language pairs. The performance of our classifier on the test sets is...

resource requirement of training corpus for low-density language pairs while the model-based pivot strategy performs worse than...

utilizing bilingual dictionaries, Penn Discourse Treebank and parallel data between English and Chinese. We start from translating...

induction, particularly in a setting where only limited parallel data are available. Word translation pairs are extracted from...corpora either do not exist or are of limited size for most language pairs. Therefore the focus of the researchers has turned towards...

machine translation. We learn rules for 8 different language pairs, showing BLEU improvements for all of them, and demonstrate...

observe the amount of symmetry in the token overlap inside language pairs. The largest difference in the token overlap is shown...identification relies on the prerequisite of possessing parallel data of closely related languages. Since parallel texts communicate...

is derived from co-occurrence statistics in bilingual parallel data. These representations can be used for a number of cr...

languages which can be translated by systems for different language pairs and whose outputs can be successfully combined into better...

reordered alternatives? Do these improvements persist across language pairs that exhibit significantly different reodering effects...

provides an abundance of readily available monolingual text, parallel data is still a comparatively scarce resource, yet plays a...

English– German and the French–Latin pairs, all remaining language pairs are quite challenging for a cognate identification program...

language models: the first on the English side of the parallel data, and the second on the English Gigaword corpus. Our...

Urdu. We trained such alignments using using GIZA++ on parallel data with partial words for Pashto sentences. The fourth type...

obtain sufficient amounts of in-domain data (in particular parallel data required for translation and distortion models) to train...Environment (env) and Labour Legislation (lab) and two language pairs: English–French and English–Greek (both directions) and...

the English–French (En–Fr) and English–German (En–De) language pairs. We use three different freely available parallel corpora...of having poor word alignments due to small amounts of parallel data in every iteration. Word-alignments are known to benefit...

hierarchical phrase based MT systems directly from the parallel data, independently of bilingual word alignments. Let us have...the behavior of the rule arithmetic on two different language pairs: German-English and Farsi-English. We also propose an...

corpora are scarce resources, especially for uncommon language pairs. Comparable corpora refer to texts that are not direct...

effective in capturing the long-range reordering between language pairs with very different word orders like Japanese-English...automatically learned from the word-aligned source-parsed parallel data and incorporated as a tree-to-string grammar for decoding...

corpus (Koehn, 2005) as our parallel corpus. It comprises parallel data from 11 languages; in this paper, we present experiments...Dutch and Italian as TLs. First, the bitexts for the language pairs under consideration (English-Dutch, English-Italian,...

candidate translation. The method was tested on two language pairs (English-French and English-German) and with a small... domain-specific bilingual lexicons when there is no parallel data available (e.g. translation memories, multilingual t...

machine translation (MT) mostly depends on the size of parallel data available for training. Although statistical MT is considered...data sets used for all our experiments represent two language pairs with parallel data of different size and type. Statistics...