Abstract | Evaluation experiments were conducted to calculate the correlation among human judgments, along with the scores produced using automatic evaluation methods for MT outputs obtained from the 12 machine translation systems in NTCIR—7. |
Experiments | These English output sentences are sentences that 12 machine translation systems in NTCIR—7 translated from 100 Japanese sentences. |
Experiments | Table 1 presents types of the 12 machine translation systems . |
Experiments | 12 machine translation systems in respective automatic evaluation methods, and “All” are the correlation coefficients using the scores of 1,200 output sentences obtained using the 12 machine translation systems . |
Introduction | High-quality automatic evaluation has become increasingly important as various machine translation systems have developed. |
Introduction | Evaluation experiments using MT outputs obtained by 12 machine translation systems in NTCIR—7(Fujii et al., 2008) demonstrate that the scores obtained using our system yield the highest correlation with the human judgments among the automatic evaluation methods in both sentence-level adequacy and fluency. |
Abstract | In this paper, we present a simple and effective method to address the issue of how to generate diversified translation systems from a single Statistical Machine Translation (SMT) engine for system combination. |
Abstract | First, a sequence of weak translation systems is generated from a baseline system in an iterative manner. |
Abstract | Then, a strong translation system is built from the ensemble of these weak translation systems . |
Background | Suppose that there are T available SMT systems {u1(/1*1), ..., uT(/1*T)}, the task of system combination is to build a new translation system v(u1(/l*1), mm?» from mm), mfg}. |
Introduction | With the emergence of various structurally different SMT systems, more and more studies are focused on combining multiple SMT systems for achieving higher translation accuracy rather than using a single translation system . |
Introduction | To reduce the burden of system development, it might be a nice way to combine a set of translation systems built from a single translation engine. |
Introduction | A key issue here is how to generate an ensemble of diversified translation systems from a single translation engine in a principled way. |
Abstract | Due to the richness of Chinese abbreviations, many of them may not appear in available parallel corpora, in which case current machine translation systems simply treat them as unknown words and leave them untranslated. |
Abstract | Our method does not require any additional annotated data other than the data that a regular translation system uses. |
Abstract | We integrate our method into a state-of-the-art baseline translation system and show that it consistently improves the performance of the baseline system on various NIST MT test sets. |
Conclusions | Our method is scalable enough to handle large amount of monolingual data, and is essentially unsupervised as it does not require any additional annotated data than the baseline translation system . |
Conclusions | We integrate our method into a state-of-the-art phrase-based baseline translation system , i.e., Moses (Koehn et al., 2007), and show that the integrated system consistently improves the performance of the baseline system on various NIST machine translation test sets. |
Unsupervised Translation Induction for Chinese Abbreviations | o Step-2: translate the list into Chinese using a baseline translation system ; |
Unsupervised Translation Induction for Chinese Abbreviations | Step-4 and -5 are natural ways to integrate the abbreviation translation component with the baseline translation system . |
Unsupervised Translation Induction for Chinese Abbreviations | However, since most of statistical translation models (Koehn et al., 2003; Chiang, 2007; Galley et al., 2006) are symmetrical, it is relatively easy to train a translation system to translate from English to Chinese, except that we need to train a Chinese language model from the Chinese monolingual data. |
Abstract | We show that combining them with word—based n—gram models in the log—linear model of a state—of—the—art statistical machine translation system leads to improvements in translation quality as indicated by the BLEU score. |
Conclusion | The experiments presented show that predictive class-based models trained using the obtained word classifications can improve the quality of a state-of-the-art machine translation system as indicated by the BLEU score in both translation tasks. |
Experiments | We use the distributed training and application infrastructure described in (Brants et al., 2007) with modifications to allow the training of predictive class-based models and their application in the decoder of the machine translation system . |
Experiments | Instead we report BLEU scores (Papineni et al., 2002) of the machine translation system using different combinations of word- and class-based models for translation tasks from English to Arabic and Arabic to English. |
Experiments | In the subsequent experiments, we use a phrase-based statistical machine translation system based on the log-linear formulation of the problem described in (Och and Ney, 2002): |
Introduction | We then show that using partially class-based language models trained using the resulting classifications together with word-based language models in a state-of-the-art statistical machine translation system yields improvements despite the very large size of the word-based models used. |
Abstract | We evaluate our procedure on translation forests from two large-scale, state-of-the-art hierarchical machine translation systems . |
Computing Feature Expectations | Forests arise naturally in chart-based decoding procedures for many hierarchical translation systems (Chiang, 2007). |
Computing Feature Expectations | generated already by the decoder of a syntactic translation system . |
Consensus Decoding Algorithms | Modern statistical machine translation systems take as input some f and score each derivation 6 according to a linear model of features: A, -6i(f, e). |
Consensus Decoding Algorithms | The distribution P(e| f) can be induced from a translation system’s features and weights by expo-nentiating with base I) to form a log-linear model: |
Experimental Results | We evaluate these consensus decoding techniques on two different full-scale state-of-the-art hierarchical machine translation systems . |
Experimental Results | SBMT is a string-to-tree translation system with rich target-side syntactic information encoded in the translation model. |
Introduction | Translation forests compactly encode distributions over much larger sets of derivations and arise naturally in chart-based decoding for a wide variety of hierarchical translation systems (Chiang, 2007; Galley et al., 2006; Mi et al., 2008; Venugopal et al., 2007). |
Introduction | In all, we show improvements of up to 1.0 BLEU from consensus approaches for state-of-the-art large-scale hierarchical translation systems . |
Experiments | These input are simplified using our simplification system namely, the DRS-SM model and the phrase-based machine translation system (Section 3.2). |
Experiments | These four sentences are directly sent to the phrase-based machine translation system to produce simplified sentences. |
Introduction | It is useful as a preprocessing step for a variety of NLP systems such as parsers and machine translation systems (Chandrasekar et al., 1996), sum-marisation (Knight and Marcu, 2000), sentence fusion (Filippova and Strube, 2008) and semantic |
Introduction | Machine Translation systems have been adapted to translate complex sentences into $nqfleones(ZhuetaL,2010;VVubbenetaL,2012; Coster and Kauchak, 2011). |
Related Work | To account for deletions, reordering and substitution, Coster and Kauchak (2011) trained a phrase based machine translation system on the PWKP corpus while modifying the word alignment output by GIZA++ in Moses to allow for null phrasal alignments. |
Related Work | (2012) use Moses and the PWKP data to train a phrase based machine translation system augmented with a post-hoc reranking procedure designed to rank the output based on their dissimilarity from the source. |
Simplification Framework | Second, the simplified sentence(s) s’ is further simplified to s using a phrase based machine translation system (PBMT+LM). |
Simplification Framework | The DRS associated with the final M-node D fin is then mapped to a simplified sentence s’fm which is further simplified using the phrase-based machine translation system to produce the final simplified sentence ssimple. |
Discussion | We also extracted the Chinese-Spanish (CS) corpus to build a standard CS translation system , which is denoted as Standard. |
Experiments | To select translation among outputs produced by different pivot translation systems , we used SVM-light (Joachins, 1999) to perform support vector regression with the linear kernel. |
Introduction | For translations from one of the systems, this method uses the outputs from other translation systems as pseudo references. |
Introduction | The advantage of our method is that we do not need to manually label the translations produced by each translation system , therefore enabling our method suitable for translation selection among any systems without additional manual work. |
Pivot Methods for Phrase-based SMT | Given a source sentence 3, we can translate it into n pivot sentences 191,192, ..., pn using a source-pivot translation system . |
Translation Selection | We propose a method to select the optimal translation from those produced by various translation systems . |
Translation Selection | For each translation, this method uses the outputs from other translation systems as pseudo references. |
Translation Selection | can easily retrain the learner under different conditions, therefore enabling our method to be applied to sentence-level translation selection from any sets of translation systems without any additional human work. |
Conclusion | We presented a translation system making use of a subcategorization database together with source-side features. |
Experiments and evaluation | We use the hierarchical translation system that comes with the Moses SMT-package and GIZA++ to compute the word alignment, using the “grow-diag-final-and” heuristics. |
Experiments and evaluation | We report results of two types of systems (table 5): first, a regular translation system built on surface forms (i.e., normal text) and second, four inflection prediction systems. |
Introduction | We first replace inflected forms by their stems or lemmas: building a translation system on a stemmed representation of the target side leads to a simpler translation task, and the morphological information contained in the source and target language parts of the translation model is more balanced. |
Previous work | Previous work has already introduced the idea of generating inflected forms as a postprocessing step for a translation system that has been stripped of (most) target-language-specific features. |
Previous work | (2010) built translation systems that predict inflected word forms based on a large array of morphological and syntactic features, obtained from both source and target side. |
Previous work | as a hierarchical machine translation system using a string-to-tree setup. |
Translation pipeline | We use a hierarchical translation system . |
Introduction | This second problem is very difficult to address with word-based translation systems , when the relevant morphological information in the target language is either nonexistent or implicitly encoded in the source language. |
Machine translation systems and data | We integrated the inflection prediction model with two types of machine translation systems : systems that make use of syntax and surface phrase-based systems. |
Machine translation systems and data | 4.1 Treelet translation system |
Machine translation systems and data | 4.2 Phrasal translation system |
Related work | Other work closely related to ours is (Toutanova and Suzuki, 2007), which uses an independently trained case marker prediction model in an English-Japanese translation system , but it focuses on the problem of generating a small set of closed class words rather |
Abstract | The experimental results show that the proposed method outperforms the baseline statistical machine translation system by 30.42%. |
Experiments | In order to evaluate the influence of segmentation results upon the statistical ON translation system , we compare the results of two translation models. |
Experiments | Then the phrase-based machine translation system MOSES2 is adopted to translate the 503 Chinese NEs in testing set into English. |
Experiments | 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 . |
Introduction | 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. |
The Framework of Our System | The Framework of our ON translation system shown in Figure 1 has four modules. |
Abstract | Parallel text is the fuel that drives modern machine translation systems . |
Abstract | Even without extensive preprocessing, the data improves translation performance on strong baseline news translation systems in five different language pairs (§4). |
Abstract | As we have shown, it is possible to obtain parallel text for many language pairs in a variety of domains very cheaply and quickly, and in sufficient quantity and quality to improve statistical machine translation systems . |
BabelNet | using (a) the human-generated translations provided in Wikipedia (the so-called inter-language links), as well as (b) a machine translation system to translate occurrences of the concepts within sense-tagged corpora, namely SemCor (Miller et al., 1993) — a corpus annotated with WordNet senses — and Wikipedia itself (Section 3.3). |
Conclusions | Further, we contribute a large set of sense occurrences harvested from Wikipedia and SemCor, a corpus that we input to a state-of-the-art machine translation system to fill in the gap between resource-rich languages — such as English — and resource-poorer ones. |
Experiment 2: Translation Evaluation | both from Wikipedia and the machine translation system . |
Methodology | Note that translations are sense-specific, as the context in which a term occurs is provided to the translation system . |
Methodology | An initial prototype used a statistical machine translation system based on Moses (Koehn et al., 2007) and trained on Europarl (Koehn, 2005). |
A Ranking Problem | For several years, WMT used the following heuristic for ranking the translation systems: |
From Rankings to Relative Ability | Ostensibly the purpose of a translation competition is to determine the relative ability of a set of translation systems . |
From Rankings to Relative Ability | Let 8 be the space of all translation systems . |
The WMT Translation Competition | Every year, the Workshop on Machine Translation (WMT) conducts a competition between machine translation systems . |
The WMT Translation Competition | The WMT organizers invite research groups to submit translation systems in eight different tracks: Czech to/from English, French to/from English, German to/from English, and Spanish to/from English. |
Abstract | We give a formal definition of one such linear-time syntactic language model, detail its relation to phrase-based decoding, and integrate the model with the Moses phrase-based translation system . |
Introduction | Bottom-up and top-down parsers typically require a completed string as input; this requirement makes it difficult to incorporate these parsers into phrase-based translation, which generates hypothesized translations incrementally, from left-to-right.1 As a workaround, parsers can rerank the translated output of translation systems (Och et al., 2004). |
Related Work | (2003) use syntactic language models to rescore the output of a tree-based translation system . |
Related Work | Post and Gildea (2009) use tree substitution grammar parsing for language modeling, but do not use this language model in a translation system . |
Related Work | The use of very large n-gram language models is typically a key ingredient in the best-performing machine translation systems (Brants et al., 2007). |
Alignment | First we obtain all models needed for a normal translations system . |
Alignment | The idea of forced alignment is to perform a phrase segmentation and alignment of each sentence pair of the training data using the full translation system as in decoding. |
Conclusion | In addition to the improved performance, the trained models are smaller leading to faster and smaller translation systems . |
Related Work | In (Liang et al., 2006) a discriminative translation system is described. |
Related Work | full and competitive translation system as starting point with reordering and all models included. |
Abstract | Our model outperforms a GIZA++ Model-4 baseline by 6.3 points in F-measure, yielding a 1.1 BLEU score increase over a state-of-the-art syntax-based machine translation system . |
Experiments | For each set of translation rules, we train a machine translation system and decode a held-out test corpus for which we report results below. |
Experiments | We use a syntax-based translation system for these experiments. |
Introduction | Automatic word alignment is generally accepted as a first step in training any statistical machine translation system . |
Introduction | Generative alignment models like IBM Model-4 (Brown et al., 1993) have been in wide use for over 15 years, and while not perfect (see Figure 1), they are completely unsupervised, requiring no annotated training data to learn alignments that have powered many current state-of-the-art translation system . |
Introduction | Instead of tasking translators to post-edit the output of machine translation systems , a more interactive approach may be more fruitful. |
Introduction | The standard approach to this problem uses the search graph of the machine translation system . |
Properties of Core Algorithm | In the project’s first field trialz, professional translators corrected machine translations of news stories from a competitive English—Spanish machine translation system (Koehn and Haddow, 2012). |
Word Completion | When the machine translation system decides for college over university, but the user types the letter u, it should change its prediction. |
Experiment | Therefore, based on this advantage, although the number of matching PASs decreases, IC-PASTR still improves the translation system using PASTR significantly. |
Integrating into the PAS-based Translation Framework | For inside context integration, since the format of IC-PASTR is the same to PASTR4, we can use the IC-PASTR to substitute PASTR for building a PAS-based translation system directly. |
Integrating into the PAS-based Translation Framework | In addition, since our method of rule extraction is different from (Zhai et al., 2012), we also use PASTR to construct a translation system as the baseline system, which we call “PASTR”. |
Introduction | Experiments show that the two PAS disambiguation methods significantly improve the baseline translation system . |
Evaluation | art machine translation system (the syntax-based variant of Joshua) achieves a score of 26.91, which is reported in (Zaidan and Callison-Burch, 2011). |
Related work | These have focused on an iterative collaboration between monolingual speakers of the two languages, facilitated with a machine translation system . |
Related work | Although hiring professional translators to create bilingual training data for machine translation systems has been deemed infeasible, Mechanical Turk has provided a low cost way of creating large volumes of translations (Callison-Burch, 2009; Ambati and Vogel, 2010). |
Related work | (2013) translated 1.5 million words of Levine Arabic and Egyptian Arabic, and showed that a statistical translation system trained on the dialect data outperformed a system trained on 100 times more MSA data. |
Abstract | This paper explores a simple and effective unified framework for incorporating soft linguistic reordering constraints into a hierarchical phrase-based translation system : l) a syntactic reordering model that explores reorderings for context free grammar rules; and 2) a semantic reordering model that focuses on the reordering of predicate-argument structures. |
Abstract | Experiments on Chinese-English translation show that the reordering approach can significantly improve a state-of-the-art hierarchical phrase-based translation system . |
Conclusion and Future Work | Experiments on Chinese-English translation show that the reordering approach can significantly improve a state-of-the-art hierarchical phrase-based translation system . |
Discussion | Then we evaluate the automatic reordering outputs generated from both our translation systems and maximum entropy classifiers. |
Experiments | Parser training includes GEOQUERY test data in order to be less dependent on parse and execution failures in the evaluation: If a translation system , response-based or reference-based, translates the German input into the gold standard English query it should be rewarded by positive task feedback. |
Experiments | We report BLEU (Papineni et al., 2001) of translation system output measured against the original English queries. |
Experiments | Furthermore, we report precision, recall, and Fl-score for executing semantic parses built from translation system outputs against the GEOQUERY database. |
Related Work | Interactive scenarios have been used for evaluation purposes of translation systems for nearly 50 years, especially using human reading comprehension testing (Pfafflin, 1965; Fuji, 1999; Jones et al., 2005), and more recently, using face-to-face conversation mediated via machine translation (Sakamoto et al., 2013). |
Abstract | Automatic word alignment is a key step in training statistical machine translation systems . |
Abstract | We propose and extensively evaluate a simple method for using alignment models to produce alignments better-suited for phrase-based MT systems, and show significant gains (as measured by BLEU score) in end-to-end translation systems for six languages pairs used in recent MT competitions. |
Word alignment results | Unfortunately, as was shown by Fraser and Marcu (2007) AER can have weak correlation with translation performance as measured by BLEU score (Pa-pineni et al., 2002), when the alignments are used to train a phrase-based translation system . |
Word alignment results | In the next section we evaluate and compare the effects of the different alignments in a phrase based machine translation system . |
Conclusion | Model MAE RMSE p 0.5596 0.7053 MA 0.5184 0.6367 us 0.5888 0.7588 MT 0.6300 0.8270 Pooled SVM 0.5823 0.7472 Independent A SVM 0.5058 0.6351 EasyAdapt SVM 0.7027 0.8816 SINGLE-TASK LEARNING Independent A 0.5091 0.6362 Independents 0.5980 0.7729 Pooled 0.5834 0.7494 Pooled & {N} 0.4932 0.6275 MULTITASK LEARNING: Annotator Combined A 0.4815 0.6174 CombinedA & {N} 0.4909 0.6268 Combined+A 0.4855 0.6203 Combined+A & {N} 0.4833 0.6102 MULTITASK LEARNING: Translation system Combineds 0.5825 0.7482 MULTITASK LEARNING: Sentence pair CombinedT 0.5813 0.7410 MULTITASK LEARNING: Combinations Combined A, 5 0.4988 0.6490 Combined A, s & {N A, 5} 0.4707 0.6003 Combined+A, 5 0.4772 0.6094 Combined 14,51 0.4588 0.5852 Combined A, s,T & {N A, 5} 0.4723 0.6023 |
Conclusion | Models of individual annotators could be used to train machine translation systems to optimise an annotator-specific quality measure, or in active learning for corpus annotation, where the model can suggest the most appropriate instances for each annotator or the best annotator for a given instance. |
Gaussian Process Regression | Let B (i) be a square covariance matrix for the ith task descriptor of M, with a column and row for each value (e. g., annotator identity, translation system , etc.). |
Introduction | We address this problem using multitask learning in which we learn individual models for each context (the task, incorporating the annotator and other metadata: translation system and the source sentence) while also modelling correlations between tasks such that related tasks can mutually inform one another. |
Conclusion and Future Work | We demonstrated that our EMBOT-based machine translation system beats a standard tree-to-tree system (Moses tree-to-tree) on the WMT 2009 translation task English —> German. |
Experiments | Our contrastive system is the 6MBOT—based translation system presented here. |
Introduction | Besides phrase-based machine translation systems (Koehn et al., 2003), syntax-based systems have become widely used because of their ability to handle nonlocal reordering. |
Introduction | In this contribution, we report on our novel statistical machine translation system that uses an [MBOT-based translation model. |
Abstract | We propose a novel approach, ensemble decoding, which combines a number of translation systems dynamically at the decoding step. |
Ensemble Decoding | The current implementation is able to combine hierarchical phrase-based systems (Chiang, 2005) as well as phrase-based translation systems (Koehn et al., 2003). |
Ensemble Decoding | However, the method can be easily extended to support combining a number of heterogeneous translation systems e.g. |
Introduction | We have modified Kriya (Sankaran et al., 2012), an in-house implementation of hierarchical phrase-based translation system (Chiang, 2005), to implement ensemble decoding using multiple translation models. |
Experiments | We trained several phrasal translation systems , varying only the word alignment (or phrasal alignment) method. |
Introduction | Most state-of—the-art statistical machine translation systems are based on large phrase tables extracted from parallel text using word-level alignments. |
Introduction | While this approach has been very successful, poor word-level alignments are nonetheless a common source of error in machine translation systems . |
Summary of the Pipeline | From this alignment, phrase pairs are extracted in the usual manner, and a phrase-based translation system is trained. |
Introduction | Recent research has shown substantial improvements can be achieved by utilizing consensus statistics obtained from outputs of multiple machine translation systems . |
Introduction | Typically, the resulting systems take outputs of individual machine translation systems as |
Introduction | A common property of all the work mentioned above is that the combination models work on the basis of n-best translation lists (full hypotheses) of existing machine translation systems . |
Abstract | This paper extends the training and tuning regime for phrase-based statistical machine translation to obtain fluent translations into morphologically complex languages (we build an English to Finnish translation system ). |
Conclusion and Future Work | In order to help with replication of the results in this paper, we have run the various morphological analysis steps and created the necessary training, tuning and test data files needed in order to train, tune and test any phrase-based machine translation system with our data. |
Experimental Results | In all the experiments conducted in this paper, we used the Moses5 phrase-based translation system (Koehn et al., 2007), 2008 version. |
Experimental Results | For evaluation against segmented translation systems in segmented forms before word reconstruction, we also segmented the baseline system’s word-based output. |
Abstract | Statistical machine translation systems combine the predictions of two directional models, typically using heuristic combination procedures like grow-diag-final. |
Experimental Results | Extraction-based evaluations of alignment better coincide with the role of word aligners in machine translation systems (Ayan and Dorr, 2006). |
Experimental Results | Finally, we evaluated our bidirectional model in a large-scale end-to-end phrase-based machine translation system from Chinese to English, based on the alignment template approach (Och and Ney, 2004). |
Introduction | Machine translation systems typically combine the predictions of two directional models, one which aligns f to e and the other e to f (Och et al., 1999). |
Experiments | We compare against a state-of-the-art hierarchical translation (Chiang, 2005) baseline, based on the Joshua translation system under the default training and decoding settings (j o sh—ba se). |
Experiments | The decoder does not employ any ‘glue grammar’ as is usual with hierarchical translation systems to limit reordering up to a certain cutoff length. |
Introduction | Interestingly, early on (Koehn et al., 2003) exemplified the difficulties of integrating linguistic information in translation systems . |
Related Work | We show that a translation system based on such a joint model can perform competitively in comparison with conditional probability models, when it is augmented with a rich latent hierarchical structure trained adequately to avoid overfitting. |
Abstract | In this paper we show how to train statistical machine translation systems on real-life tasks using only nonparallel monolingual data from two languages. |
Conclusion | This work serves as a big step towards large-scale unsupervised training for statistical machine translation systems . |
Experimental Evaluation | Och (2002) reports results of 48.2 BLEU for a single-word based translation system and 56.1 BLEU using the alignment template approach, both trained on parallel data. |
Related Work | Unsupervised training of statistical translations systems without parallel data and related problems have been addressed before. |
Experiments | (2004) and Cherry and Quirk (2008) both use the l-best output of a machine translation system . |
Experiments | Cherry and Quirk (2008) report an accuracy of 71.9% on a similar experiment with German a source language, though the translation system and training data were different so the numbers are not comparable. |
Experiments | (2004) and Cherry and Quirk (2008) in evaluating our language models on their ability to distinguish the l-best output of a machine translation system from a reference translation in a pairwise fashion. |
Introduction | N -gram language models are a central component of all speech recognition and machine translation systems , and a great deal of research centers around refining models (Chen and Goodman, 1998), efficient storage (Pauls and Klein, 2011; Heafield, 2011), and integration into decoders (Koehn, 2004; Chiang, 2005). |
Abstract | Typical statistical machine translation systems are batch trained with a given training data and their performances are largely influenced by the amount of data. |
Introduction | Most of them have been proposed in order to make translation systems perform better for resource-scarce domains when most training data comes from resource-rich domains, and ignore performance on a more generic domain without domain bias (Wang et al., 2012). |
Related Work | Bilingual phrases are cornerstones for phrase-based SMT systems (Och and Ney, 2004; Koehn et al., 2003; Chiang, 2005) and existing translation systems often get ‘crowd-sourced’ improvements (Levenberg et al., 2010). |
Abstract | Two decades after their invention, the IBM word-based translation models, widely available in the GIZA++ toolkit, remain the dominant approach to word alignment and an integral part of many statistical translation systems . |
Conclusion | We hope that our method, due to its simplicity, generality, and effectiveness, will find wide application for training better statistical translation systems . |
Experiments | We then tested the effect of word alignments on translation quality using the hierarchical phrase-based translation system Hiero (Chiang, 2007). |
A semantic span can include one or more eus. | Most translation systems adopt the features from a translation model, a language model, and sometimes a reordering model. |
Abstract | The two models are integrated into a hierarchical phrase-based translation system to evaluate their effectiveness. |
Experiments | First, we adopted only the tagged-flattened rules in the hierarchical translation system . |
Abstract | In this paper we study the use of sentence-level dialect identification in optimizing machine translation system selection when translating mixed dialect input. |
Abstract | We test our approach on Arabic, a prototypical diglossic language; and we optimize the combination of four different machine translation systems . |
Machine Translation Experiments | We use the open-source Moses toolkit (Koehn et al., 2007) to build four Arabic-English phrase-based statistical machine translation systems (SMT). |
Alternatives to Correlation-based Meta-evaluation | The translation system obviates some information which, in context, is not considered crucial by the human assessors. |
Alternatives to Correlation-based Meta-evaluation | We consider the set of translations system presented in each competition as the translation approaches pool. |
Conclusions | In addition, our Combined System Test shows that, when training a combined translation system , using metrics at several linguistic processing levels improves substantially the use of individual metrics. |
Abstract | Modern phrase-based machine translation systems make extensive use of word-based translation models for inducing alignments from parallel corpora. |
Experiments | 10Using the factorised alignments directly in a translation system resulted in a slight loss in BLEU versus using the un-factorised alignments. |
Introduction | Leading translation systems (Chiang, 2007; Koehn et al., 2007; Marcu et al., 2006) all use some kind of multi-word translation unit, which allows translations to be produced from large canned units of text from the training corpus. |
Experiments | We use the NiuTrans 2 toolkit which adopts GIZA++ (Och and Ney, 2003) and MERT (Och, 2003) to train and tune the machine translation system . |
Experiments | This tool scores the outputs in several criterions, while the case-insensitive BLEU-4 (Papineni et al., 2002) is used as the evaluation for the machine translation system . |
Experiments | When top 600k sentence pairs are picked out from general-domain corpus to train machine translation systems , the systems perform higher than the General-domain baseline trained on 16 million parallel data. |
Building the Corpus | In addition, the overall measure of success—induction of machine translation systems from limited resources—pushes the state of the art (Kumar et al., 2007). |
Conclusion | We need leaner methods for building machine translation systems ; new algorithms for cross-linguistic bootstrapping via multiple paths; more effective techniques for leveraging human effort in labeling data; scalable ways to get bilingual text for unwritten languages; and large scale social engineering to make it all happen quickly. |
Human Language Project | Another layer of the corpus consists of sentence and word alignments, required for training and evaluating machine translation systems , and for extracting bilingual lexicons. |
Abstract | We propose to incorporate two groups of linguistic features, which convey information from outside machine translation systems , into error detection: lexical and syntactic features. |
Conclusions and Future Work | Therefore our approach can be used for other machine translation systems , such as rule-based or example-based system, which generally do not produce N -best lists. |
SMT System | To obtain machine-generated translation hypotheses for our error detection, we use a state-of-the-art phrase-based machine translation system MOSES (Koehn et al., 2003; Koehn et al., 2007). |
Name-aware MT | Then we apply a state-of-the-art name translation system (Ji et al., 2009) to translate names into the target language. |
Name-aware MT | The name translation system is composed of the following steps: (1) Dictionary matching based on 150,041 name translation pairs; (2) Statistical name transliteration based on a structured perceptron model and a character based MT model (Dayne and Shahram, 2007); (3) Context information extraction based re-ranking. |
Name-aware MT | For those names with fewer than five instances in the training data, we use the name translation system to provide translations; for the rest of the names, we leave them to the baseline MT model to handle. |
Abstract | Most modern machine translation systems use phrase pairs as translation units, allowing for accurate modelling of phrase-internal translation and reordering. |
Model | We consider a process in which the target string is generated using a left-to-right order, similar to the decoding strategy used by phrase-based machine translation systems (Koehn et al., 2003). |
Related Work | (2011) develop a bilingual language model which incorporates words in the source and target languages to predict the next unit, which they use as a feature in a translation system . |
Abstract | Preordering of a source language sentence to match target word order has proved to be useful for improving machine translation systems . |
Experimental setup | The parallel corpus is used for building our phrased based machine translation system and to add training data for our reordering model. |
Introduction | Dealing with word order differences between source and target languages presents a significant challenge for machine translation systems . |
Related Work | They use separate translation systems for each domain, and a supervised setting, whereas we aim for a system that integrates support for multiple domains, with or without supervision. |
Translation Model Architecture | ment a multi-domain translation system . |
Translation Model Architecture | The translation model framework could also serve as the basis of real-time adaptation of translation systems , e. g. by using incremental means to update the weight vector, or having an incrementally trainable component model that learns from the post-edits by the user, and is assigned a suitable weight. |
Conclusions and Future Work | We have produced initial results in terms of rule extraction, and we will be integrating these rules into the full Italian-LIS translation system to produce improved translation of connec-t1ves. |
Introduction | The resulting lack of a shared written form does nothing to improve the availability of sign language corpora; bilingual corpora, which are of particular importance to a translation system , are especially rare. |
The effect of the Italian connectives on the LIS translation | Tree alignment in a variety of forms has been extensively used in machine translation systems (Gildea, 2003; Eisner, 2003; May and Knight, 2007). |
Experiments | well handled by all machine translation systems 2. |
Experiments | 2We tested other translation systems , but Google Translate gave the best results. |
Introduction | One of the questions posed was whether the quality of present machine translation systems would enable to learn the classification properly. |