This paper introduces a novel pre-ordering approach based on dependency parsing for Chinese-English SMT.

In this paper, we introduced a novel pre-ordering approach based on dependency parsing for a Chinese-English PBSMT system.

These results indicated that dependency parsing is more effective for conducting pre-ordering for Chinese-English PBSMT.

Our development set was the official NIST MT evaluation data from 2002 to 2005, consisting of 4476 Chinese-English sentences pairs.

Syntax-based pre-ordering by employing constituent parsing have demonstrated effectiveness in many language pairs, such as English-French (Xia and McCord, 2004), German-English (Collins et al., 2005), Chinese-English (Wang et al., 2007; Zhang et al., 2008), and English-Japanese (Lee et al., 2010).

The purpose of this paper is to introduce a novel dependency-based pre-ordering approach through creating a pre-ordering rule set and applying it to the Chinese-English PBSMT system.

To our knowledge, our manually created pre-ordering rule set is the first Chinese-English dependency-based pre-ordering rule set.

We achieve significant improvement over both Hiero and phrase-based baselines for Arabic-English, Chinese-English and German-English translation.

We will report the impact of integrating phrase-based features into Hiero systems for three language pairs: Arabic-English, Chinese-English and German-English.

4.3 Chinese-English Results

The Chinese-English system was trained on FBIS corpora of 384K sentence pairs, the English corpus is lower case.

In our Chinese-English experiment, the Hiero system still outperforms the discontinuous phrase-based system.

(2008) added structure distortion features into their decoder and showed improvements in their Chinese-English experiment.

In the experiment section, we will discuss the impact of removing rules with nonaligned sub-phrases in our German-English and Chinese-English experiments.

We make a detailed analysis on the Chinese-English translation results that are affected by our paraphrase rules.

The analysis is carried out on the IWSLT 2007 Chinese-English test set, 84 out of 489 input sentences have been affected by paraphrases, and the statistic of human evaluation is shown in Table 8.

The experiments were conducted in both Chinese-English and English-Chinese directions for the oral group, and Chinese-English direction for the news group.

corpora, including the Chinese-English Sentence Aligned Bilingual Corpus (CLDC-LAC-2003-004) and the Chinese-English Parallel Corpora (CLDC-LAC-2003-006).

For testing and developing, we used six Chinese-English development corpora of IWSLT 2008.

We also show strong improvements on the NIST OpenMT12 Chinese-English task, as well as the DARPA BOLT (Broad Operational Language Translation) Arabic-English and Chinese-English conditions.

We present MT primary results on Arabic-English and Chinese-English for the NIST OpenMT12 and DARPA BOLT conditions.

Table 3: Primary results on Arabic-English and Chinese-English NIST MT12 Test Set.

For the Chinese-English condition, there is an improvement of +0.8 BLEU from the primary NNJM and +1.3 BLEU overall.

An example of the NNJ M context model for a Chinese-English parallel sentence is given in Figure 1.

We evaluate our optimizer on Chinese-English and Arabic-English translation tasks, each with small and large feature sets, and show that our learner is able to achieve significant improvements of 1.2-2 BLEU and 1.7-4.3 TER on average over state-of-the-art optimizers with the large feature set.

In both Arabic-English feature sets, MIRA seems to take the second place, while RAMPION lags behind, unlike in Chinese-English (§4).6

Spread analysis: For RM, the average spread of the projected data in the Chinese-English small feature set was 0.9i3.6 for all tuning iterations, and 0.7j:2.9 for the iteration with the highest decoder performance.

To evaluate the advantage of explicitly accounting for the spread of the data, we conducted several experiments on two Chinese-English translation test sets, using two different feature sets in each.

As can be seen from the results in Table 3, our RM method was the best performer in all Chinese-English tests according to all measures — up to 1.9 BLEU and 6.6 TER over MIRA — even though we only optimized for BLEU.5 Surprisingly, it seems that MIRA did not benefit as much from the sparse features as RM.

In preliminary experiments with a smaller trigram LM, our RM method consistently yielded the highest scores in all Chinese-English tests — up to 1.6 BLEU and 6.4 TER from MIRA, the second best performer.

Chinese-English translation experiments show that our algorithm, RM, significantly outperforms strong state-of-the-art optimizers, in both a basic feature setting and high-dimensional (sparse) feature space (§4).

Additionally, we remove low confidence alignment links from the word alignment of a bilingual training corpus, which increases the alignment F-score, improves Chinese-English and Arabic-English translation quality and significantly reduces the phrase translation table size.

lected among multiple alignments and it obtained 0.8 F-measure improvement over the single best Chinese-English aligner.

When we removed low confidence links from the MaXEnt aligner, we reduced the Chinese-English alignment error by 5% and the Arabic-English alignment error by 10%.

We applied the confidence-based link filtering on Chinese-English and Arabic-English word alignment.

We randomly selected 512 Chinese-English (CE) sentence pairs and generated word alignment using the MaxEnt aligner (Ittycheriah and Roukos, 2005).

We evaluate the improved alignment on several Chinese-English and Arabic-English machine translation tasks.

In the Chinese-English MT experiment, we selected 40 NW documents, 41 WE documents as the test set, which includes 623 sentences with 16667 words.

For our experiments, we tuned a on Chinese-English data, setting it to 0.25 and keeping this value for the other language pairs.

The large data condition uses training data from NIST3 2009 ( Chinese-English track).

We are currently investigating why PORT tuning gives higher BLEU scores than BLEU tuning for Chinese-English and German-English.

PORT outperforms Qmean on seven of the eight automatic scores shown for small and large Chinese-English .

Experiments on Chinese-English translation demonstrated the effectiveness of our approach on enhancing the quality of overall translation, name translation and word alignment over a high-quality MT baselinel.

As our baseline, we apply a high-performing Chinese-English MT system (Zheng, 2008; Zheng et al., 2009) based on hierarchical phrase-based translation framework (Chiang, 2005).

Experiments on Chinese-English translation demonstrated the effectiveness of our approach over a high-quality MT baseline in both overall translation and name translation, especially for formal genres.

We used a large Chinese-English MT training corpus from various sources and genres (including newswire, web text, broadcast news and broadcast conversations) for our experiments.

get side of Chinese-English and Egyptian Arabic-English parallel text, English monolingual discussion forums data Rl-R4 released in BOLT Phase 1 (LDC2012E04, LDC2012E16, LDC2012E21, LDC2012E54), and English Gigaword Fifth Edition (LDC2011T07).

We conducted the experiment on the Chinese-English Parallel Treebank (Li et a1., 2010) with ground-truth word alignment.

To demonstrate the effect of the {O-norm on the IBM models, we performed experiments on four translation tasks: Arabic-English, Chinese-English , and Urdu-English from the NIST Open MT Evaluation, and the Czech-English translation from the Workshop on Machine Translation (WMT) shared task.

0 Chinese-English : selected data from the constrained task of the NIST 2009 Open MT Evaluation.3

For Arabic-English and Chinese-English , we used 346 and 184 hand-aligned sentences from LDC2006E86 and LDC2006E93.

We have been able to extract over 1M Chinese-English parallel segments from Sina Weibo (the Chinese counterpart of Twitter) using only their public APIs.

For the news test, we created a new test set from a crawl of the Chinese-English documents on the Project Syndicate website2, which contains news commentary articles.

Second, we use the full 2012 NIST Chinese-English dataset (approximately 8M sentence pairs, including FBIS).

Section 5 describes the data we gathered from both Sina Weibo (Chinese-English) and Twitter ( Chinese-English and Arabic-English).

The target domains in this work are Twitter and Sina Weibo, and the main language pair is Chinese-English .

This means that for the Chinese-English language pair, we only keep tweets with more than 3 Mandarin characters and 3 latin words.

Speed Ratio 29 142 Chinese-English Objective Hiero SBMT

We evaluated on both Chinese-English and Arabic-English translation tasks.

For the Chinese-English experiments, we used 260 million words of word-aligned parallel text; the hierarchical system used all of this data, and the syntax-based system used a 65-million word subset.

As our approach combines the merits of phrase-based and string-to-dependency models, it achieves significant improvements over the two baselines on the NIST Chinese-English datasets.

We evaluate our method on the NIST Chinese-English translation datasets.

1Empirically, we find that the average number of stacks for J words is about 1.5 X J on the Chinese-English data.

We evaluated our phrase-based string-to-dependency translation system on Chinese-English translation.

We can’t get Chinese-English bilingual pages when the input is a Chinese query.

We use syllables as translation units to build a statistical Chinese-English backward transliteration model in our system.

Based on the above alignment method, we can get our statistical Chinese-English backward transliteration model as,

Chinese-English backward transliteration has some differences from traditional translation.

The first experiments are on the IWSLT data set for Chinese-English translation.

Table 3: Machine translation performance in BLE U % on the IWSLT 2005 Chinese-English test set.

To test whether our improvements carry over to larger datasets, we assess the performance of our model on the FBIS Chinese-English data set.

We demonstrate our model on Chinese-English and Arabic-English translation datasets.

On two Chinese-English tasks, our semi-supervised DAE features obtain statistically significant improvements of l.34/2.45 (IWSLT) and 0.82/1.52 (NIST) BLEU points over the unsupervised DBN features and the baseline features, respectively.

We now test our DAE features on the following two Chinese-English translation tasks.

The bilingual corpus is the Chinese-English part of Basic Traveling Expression corpus (BTEC) and China-Japan-Korea (CJK) corpus (0.38M sentence pairs with 3.5/3.8M Chi-nese/English words).

Finally, we conduct large-scale experiments on IWSLT and NIST Chinese-English translation tasks, respectively, and the results demonstrate that our solutions solve the two aforementioned shortcomings successfully.

In this work we consider two types of code-switched documents: single messages and conversations, and two language pairs: Chinese-English and Spanish-English.

An example of a Chinese-English code-switched messages is given by Ling et al.

We used two datasets: a Sina Weibo Chinese-English corpus (Ling et al., 2013) and a Spanish-English Twitter corpus.

We apply our approach to a state-of-the-art phrase-based system and demonstrate very promising BLEU improvements and TER reductions on the NIST Chinese-English MT evaluation data.

We experimented with our approach on Chinese-English translation using the NiuTrans open-source MT toolkit (Xiao et al., 2012).

It contains the annotation of sentence skeleton on the Chinese-language side of the Penn Parallel Chinese-English Treebank (LD-C2003E07).

0 We apply the proposed model to Chinese-English phrase-based MT and demonstrate promising BLEU improvements and TER reductions on the NIST evaluation data.

The resulting predictions improve the precision and recall of both alignment links and extraced phrase pairs in Chinese-English experiments.

We evaluated alignment quality on a hand-aligned portion of the NIST 2002 Chinese-English test set (Ayan and Dorr, 2006).

AER results for Chinese-English are reported in Table 2.

Our model-based approach to aligner combination yields improvements in alignment quality and phrase extraction quality in Chinese-English experiments, relative to typical heuristic combinations methods applied to the predictions of independent directional models.

Experimental results on the NIST MT-2003 Chinese-English translation task show that our method statistically significantly outperforms the four baseline systems.

Finally, we examine our methods on the FBIS corpus and the NIST MT-2003 Chinese-English translation task.

We evaluate our method on Chinese-English translation task.

We evaluate our method on the NIST MT-2003 Chinese-English translation tasks.

It proves that our system can work well on the Chinese-English ON translation task.

Compared with the statistical ON translation model, we can see that the performance is improved from 18.29% to 48.71% (the bold data shown in column 1 and column 3 of Table 5) by using our Chinese-English ON translation system.

In order to use the web information to assist Chinese-English ON translation, we must firstly retrieve the bilingual web pages effectively.

For solving these two problems, we propose a Chinese-English organization name translation system using heuristic web mining and asymmetric alignment, which has three innovations.

Table 2 describes the data used for model training in this paper, including the BTEC (Basic Travel Expression Corpus) Chinese-English (CE) corpus and the BTEC English-Spanish (ES) corpus provided by IWSLT 2008 organizers, the HIT olympic CE corpus (2004-863-008)1 and the Europarl ES corpusz.

For Chinese-English translation, we mainly used BTEC CE1 corpus.

We used two commercial RBMT systems in our experiments: System A for Chinese-English bidirectional translation and System B for English-Chinese and English-Spanish translation.

Our string-to-dependency system generates 80% fewer rules, and achieves 1.48 point improvement in BLEU and 2.53 point improvement in TER on the decoding output on the NIST 04 Chinese-English evaluation set.

We used part of the NIST 2006 Chinese-English large track data as well as some LDC corpora collected for the DARPA GALE program (LDC2005E83, LDC2006E34 and LDC2006G05) as our bilingual training data.

For example, Chiang (2007) showed that the Hiero system achieved about 1 to 3 point improvement in BLEU on the NIST 03/04/05 Chinese-English evaluation sets compared to a start-of-the-art phrasal system.

Our string-to-dependency decoder shows 1.48 point improvement in BLEU and 2.53 point improvement in TER on the NIST 04 Chinese-English MT evaluation set.

Large-scale experiments on Arabic-English and Chinese-English show that our method produces significant translation quality gains by exploiting sparse features.

We built Arabic-English and Chinese-English MT systems with Phrasal (Cer et al., 2010), a phrase-based system based on alignment templates (Och and Ney, 2004).

We conduct large-scale translation quality experiments on Arabic-English and Chinese-English .

In our experiments, our model improved 2.9 BLEU points for J apanese-English and 2.6 BLEU points for Chinese-English translation compared to the lexical reordering models.

Japanese-English Chinese-English HIER 30.47 32.66

Experiments confirmed the effectiveness of our method for J apanese-English and Chinese-English translation, using NTCIR-9 Patent Machine Translation Task data sets (Goto et al., 2011).

Experimental results on Chinese-English translation across four test sets demonstrate significant improvements of the HD-HPB model over both Chiang’s HPB and a source-side SAMT—style refined version of HPB.

Figure 1: An example word alignment for a Chinese-English sentence pair with the dependency parse tree for the Chinese sentence.

Experiments on Chinese-English translation using four NIST MT test sets show that our HD-HPB model significantly outperforms Chiang’s HPB as well as a SAMT—style refined version of HPB.

The system is trained on 10 million parallel sentences that are available to the Phase 1 of the DARPA BOLT Chinese-English MT task.

For our Chinese-English experiments, we use a simple heuristic that equates as anchors, single-word chunks whose corresponding word class belongs to closed-word classes, bearing a close resemblance to (Setiawan et al., 2007).

Figure 1: An aligned Chinese-English sentence pair.

Chinese-English experimental results on four typical attributes showed that WikiCiKE significantly outperforms both the current translation based methods and the monolingual extraction methods.

In this section, we present our experiments to evaluate the effectiveness of WikiCiKE, where we focus on the Chinese-English case; in other words, the target language is Chinese and the source language is English.

Chinese-English experiments for four typical attributes demonstrate that WikiCiKE outperforms both the monolingual extraction method and current translation-based method.

Experiments on Chinese-English translation show that the reordering approach can significantly improve a state-of-the-art hierarchical phrase-based translation system.

Experiments on Chinese-English translation show that the reordering approach can significantly improve a state-of-the-art hierarchical phrase-based translation system.

In this section, we test its effectiveness in Chinese-English translation.

Experimental results on the NIST MT-2005 Chinese-English translation task show that our method statistically significantly outperforms the baseline systems.

The experimental results on the NIST MT-2005 Chinese-English translation task demonstrate the effectiveness of the proposed model.

Experiment results on the NIST MT-2005 Chinese-English translation task show that our method significantly outperforms Moses (Koehn et al., 2007), a state-of-the-art phrase-based SMT system, and other linguistically syntax-based methods, such as SCFG-based and STSG-based methods (Zhang et al., 2007).

Our experimental results on IWSLT Chinese-English corpus have demonstrated consistent and significant improvement over the widely used word alignment matrix based extraction method.

We do experiments on IWSLT (Paul, 2006) 2006 Chinese-English corpus.

The training corpus consists of 40K Chinese-English parallel sentences in travel domain with to-