Phrase-based Machine Translation Research Articles

Sentiment induced phrase-based machine translation: Robustness analysis of PBSMT with senti-module

Every type of machine translation system (i.e. neural, statistical, rule-based machine translation system) is equal important to build a sophistical hybrid machine translation system. Keeping this fact in my mind, I concentrate to improve statistical machine translation system with more natural way. In this paper, I try to preserve sentiment after translation to improve the overall accuracy of the machine translation system. So, I introduced senti-model here. A senti-model (sentiment model), translation model, language model, and distortion model are incorporated on the top of the beam search algorithm for decoding. At first, sentiment information is learned and modeled with translation probability by using this algorithm. Thereafter, I decode the source sentences-based on the contextual information. Overall procedure of translation modeling with a sentiment, parameter estimation for it, and senti-translation decoding (decoding with the sentiment model) are presented with empirical evidence. Experiments on a benchmark English–Hindi dataset shows that the proposed model is capable to improve the accuracy (in terms of 4.66 BLEU points, 4.09 LeBleu points, 4.67 NIST points, 5.71 RIBES points) significantly and preserves sentiment 7.79% more than the state-of-the-art technique.

Design and Testing of Automatic Machine Translation System Based on Chinese-English Phrase Translation

With the development of linguistics and the improvement of computer performance, the effect of machine translation is getting better and better, and it is widely used. The automatic expression translation method based on the Chinese-English machine takes short sentences as the basic translation unit and makes full use of the order of short sentences. Compared with word-based statistical machine translation methods, the effect is greatly improved. The performance of machine translation is constantly improving. This article aims to study the design of phrase-based automatic machine translation systems by introducing machine translation methods and Chinese-English phrase translation, explore the design and testing of machine automatic translation systems based on the combination of Chinese-English phrase translation, and explain the role of machine automatic translation in promoting the development of translation. In this article, through the combination of machine translation experiments and machine automatic translation system design methods, the design and testing of machine automatic translation systems based on Chinese-English phrase translation combinations are studied to cultivate people's understanding of language, knowledge, and intelligence and then help solve other problems. Language processing issues promote the development of corpus linguistics. The experimental results in this article show that when the Chinese-English phrase translation probability table is changed from 82% to 51%, the BLEU translation evaluation system for the combination of Chinese-English phrases is improved. Automatic machine translation saves time and energy of translation work, which shows that machine translation shows its advantages due to its short development cycle and easy processing of large-scale corpora.

Unsupervised Statistical Text Simplification

Most recent approaches for Text Simplification (TS) have drawn on insights from machine translation to learn simplification rewrites from the monolingual parallel corpus of complex and simple sentences, yet their effectiveness strongly relies on large amounts of parallel sentences. However, there has been a serious problem haunting TS for decades, that is, the availability of parallel TS corpora is scarce or not fit for the learning task. In this paper, we will focus on one especially useful and challenging problem of unsupervised TS without a single parallel sentence. To the best of our knowledge, we present the first unsupervised text simplification system based on phrase-based machine translation system, which leverages a careful initialization of phrase tables and language models. On the widely used WikiLarge and WikiSmall benchmarks, our system respectively obtains 39.08 and 25.12 SARI points, even outperforms some supervised baselines.

Adaptation in Statistical Machine Translation for Low-resource Domains in English-Vietnamese Language

Adaptation in Statistical Machine Translation for Low-resource Domains in English-Vietnamese Language

Improving neural machine translation through phrase-based soft forced decoding

Compared to traditional statistical machine translation (SMT), such as phrase-based machine translation (PBMT), neural machine translation (NMT) often sacrifices adequacy for the sake of fluency. We propose a method to combine the advantages of traditional SMT and NMT by exploiting an existing phrase-based SMT model to compute the phrase-based decoding cost for an NMT output and then using this cost to rerank the n-best NMT outputs. The main challenge in implementing this approach is that NMT outputs may not be in the search space of the standard phrase-based decoding algorithm, because the search space of PBMT is limited by the phrase-based translation rule table. We propose a phrase-based soft forced decoding algorithm, which can always successfully find a decoding path for any NMT output. We show that using the phrase-based decoding cost to rerank the NMT outputs can successfully improve translation quality on four different language pairs.

Leveraging Additional Resources for Improving Statistical Machine Translation on Asian Low-Resource Languages

Phrase-based machine translation (MT) systems require large bilingual corpora for training. Nevertheless, such large bilingual corpora are unavailable for most language pairs in the world, causing a bottleneck for the development of MT. For the Asian language pairs—Japanese, Indonesian, Malay paired with Vietnamese—they are also not excluded from the case, in which there are no large bilingual corpora on these low-resource language pairs. Furthermore, although the languages are widely used in the world, there is no prior work on MT, which causes an issue for the development of MT on these languages. In this article, we conducted an empirical study of leveraging additional resources to improve MT for the Asian low-resource language pairs: translation from Japanese, Indonesian, and Malay to Vietnamese. We propose an innovative approach that lies in two strategies of building bilingual corpora from comparable data and phrase pivot translation on existing bilingual corpora of the languages paired with English. Bilingual corpora were built from Wikipedia bilingual titles to enhance bilingual data for the low-resource languages. Additionally, we introduced a combined model of the additional resources to create an effective solution to improve MT on the Asian low-resource languages. Experimental results show the effectiveness of our systems with the improvement of +2 to +7 BLEU points. This work contributes to the development of MT on low-resource languages, especially opening a promising direction for the progress of MT on the Asian language pairs.

Swarm Algorithms for NLP - The Case of Limited Training Data

Abstract The present article describes a novel phrasing model which can be used for segmenting sentences of unconstrained text into syntactically-defined phrases. This model is based on the notion of attraction and repulsion forces between adjacent words. Each of these forces is weighed appropriately by system parameters, the values of which are optimised via particle swarm optimisation. This approach is designed to be language-independent and is tested here for different languages. The phrasing model’s performance is assessed per se, by calculating the segmentation accuracy against a golden segmentation. Operational testing also involves integrating the model to a phrase-based Machine Translation (MT) system and measuring the translation quality when the phrasing model is used to segment input text into phrases. Experiments show that the performance of this approach is comparable to other leading segmentation methods and that it exceeds that of baseline systems.

Post-editing neural machine translation versus phrase-based machine translation for English–Chinese

This paper aims to shed light on the post-editing process of the recently-introduced neural machine translation (NMT) paradigm. Using simple and more complex texts, we first evaluate the output quality from English to Chinese phrase-based statistical (PBSMT) and NMT systems. Nine raters assess the MT quality in terms of fluency and accuracy and find that NMT produces higher-rated translations than PBSMT for both texts. Then we analyze the effort expended by 68 student translators during HT and when post-editing NMT and PBSMT output. Our measures of post-editing effort are all positively correlated for both NMT and PBSMT post-editing. Our findings suggest that although post-editing output from NMT is not always significantly faster than post-editing PBSMT, it significantly reduces the technical and cognitive effort. We also find that, in contrast to HT, post-editing effort is not necessarily correlated with source text complexity.

Efficient Phrase Table pruning for Hindi to English machine translation through syntactic and marker-based filtering and hybrid similarity measurement

AbstractFor a given training corpus of parallel sentences, the quality of the output produced by a translation system relies heavily on the underlying similarity measurement criteria. A phrase-based machine translation system derives its output through a generative process using a Phrase Table comprising source and target language phrases. As a consequence, the more effective the Phrase Table is, in terms of its size and the output that may be derived out of it, the better is the expected outcome of the underlying translation system. However, finding the most similar phrase(s) from a given training corpus that can help generate a good quality translation poses a serious challenge. In practice, often there are many parallel phrase entries in a Phrase Table that are either redundant, or do not contribute to the translation results effectively. Identifying these candidate entries and removing them from the Phrase Table will not only reduce the size of the Phrase Table, but should also help in improving the processing speed for generating the translations. The present paper develops a scheme based on syntactic structure and the marker hypothesis (Green 1979, The necessity of syntax markers: two experiments with artificial languages, Journal of Verbal Learning and Behavior) for reducing the size of a Phrase Table, without compromising much on the translation quality of the output, by retaining the non-redundant and meaningful parallel phrases only. The proposed scheme is complemented with an appropriate similarity measurement scheme to achieve maximum efficiency in terms of BLEU scores. Although designed for Hindi to English machine translation, the overall approach is quite general, and is expected to be easily adaptable for other language pairs as well.

Sentence Selection and Weighting for Neural Machine Translation Domain Adaptation

Neural machine translation (NMT) has been prominent in many machine translation tasks. However, in some domain-specific tasks, only the corpora from similar domains can improve translation performance. If out-of-domain corpora are directly added into the in-domain corpus, the translation performance may even degrade. Therefore, domain adaptation techniques are essential to solve the NMT domain problem. Most existing methods for domain adaptation are designed for the conventional phrase-based machine translation. For NMT domain adaptation, there have been only a few studies on topics such as fine tuning, domain tags, and domain features. In this paper, we have four goals for sentence level NMT domain adaptation. First, the NMT's internal sentence embedding is exploited and the sentence embedding similarity is used to select out-of-domain sentences that are close to the in-domain corpus. Second, we propose three sentence weighting methods, i.e., sentence weighting, domain weighting, and batch weighting, to balance the data distribution during NMT training. Third, in addition, we propose dynamic training methods to adjust the sentence selection and weighting during NMT training. Fourth, to solve the multidomain problem in a real-world NMT scenario where the domain distributions of training and testing data often mismatch, we proposed a multidomain sentence weighting method to balance the domain distributions of training data and match the domain distributions of training and testing data. The proposed methods are evaluated in international workshop on spoken language translation (IWSLT) English-to-French/German tasks and a multidomain English-to-French task. Empirical results show that the sentence selection and weighting methods can significantly improve the NMT performance, outperforming the existing baselines.

Language-related issues for NMT and PBMT for English–German and English–Serbian

This work presents an extensive comparison of language-related problems for neural machine translation (NMT) and phrase-based machine translation (PBMT) for German-to-English, English-to-German and English-to-Serbian. The explored issues are related both to the characteristics of the languages as well as to the (machine) translation process and, although related, go beyond typical translation error classes. It is shown that the main advantage of the NMT approach consists of better generating verb forms, avoiding verb omissions, as well as better handling of English noun collocations and negation. It is also shown that the main obstacles for the NMT system are prepositions, translation of English (source) ambiguous words and generating English (target) continuous and perfect tenses. In addition, preliminary experiments show that a number of issues are complementary, i.e., not occurring in the same segments and/or in the same form. This means that a combination or hybridisation of the NMT and PBMT approaches is a promising direction for improving both types of systems.

Quantitative fine-grained human evaluation of machine translation systems: a case study on English to Croatian

This paper presents a quantitative fine-grained manual evaluation approach to comparing the performance of different machine translation (MT) systems. We build upon the well-established Multidimensional Quality Metrics (MQM) error taxonomy and implement a novel method that assesses whether the differences in performance for MQM error types between different MT systems are statistically significant. We conduct a case study for English-to-Croatian, a language direction that involves translating into a morphologically rich language, for which we compare three MT systems belonging to different paradigms: pure phrase-based, factored phrase-based and neural. First, we design an MQM-compliant error taxonomy tailored to the relevant linguistic phenomena of Slavic languages, which made the annotation process feasible and accurate. Errors in MT outputs were then annotated by two annotators following this taxonomy. Subsequently, we carried out a statistical analysis which showed that the best-performing system (neural) reduces the errors produced by the worst system (pure phrase-based) by more than half (54\%). Moreover, we conducted an additional analysis of agreement errors in which we distinguished between short (phrase-level) and long distance (sentence-level) errors. We discovered that phrase-based MT approaches are of limited use for long distance agreement phenomena, for which neural MT was found to be especially effective.

Improve Mongolian-Chinese translation by Introducing SMT Information into NMT

In this paper, we improved the final accuracy of statistic alignment of phrase-based statistic machine translation (PBSMT) and introduced the improved alignment result into NMT. Moreover, we investigated syntax reordering and data augmentation for improving SMT alignments, which eventually leads to better NMT performance. The experiment results show that applying our approach to Mongolian-Chinese translation demonstrates promising improvements. With annotation alignment and compared to NMT baseline, we obtained up to 2.98 BLEU improvement on the low-resource language pair Mongolian-Chinese.

A survey of domain adaptation for statistical machine translation

Differences in domains of language use between training data and test data have often been reported to result in performance degradation for phrase-based machine translation models. Throughout the past decade or so, a large body of work aimed at exploring domain-adaptation methods to improve system performance in the face of such domain differences. This paper provides a systematic survey of domain-adaptation methods for phrase-based machine-translation systems. The survey starts out with outlining the sources of errors in various components of phrase-based models due to domain change, including lexical selection, reordering and optimization. Subsequently, it outlines the different research lines to domain adaptation in the literature, and surveys the existing work within these research lines, discussing how these approaches differ and how they relate to each other.

Neural versus phrase-based MT quality: An in-depth analysis on English–German and English–French

Within the field of statistical machine translation, the neural approach (NMT) is currently pushing ahead the state of the art performance traditionally achieved by phrase-based approaches (PBMT), and is rapidly becoming the dominant technology in machine translation. Indeed, in the last IWSLT and WMT evaluation campaigns on machine translation, NMT outperformed well established state-of-the-art PBMT systems on many different language pairs. To understand in what respects NMT provides better translation quality than PBMT, we perform a detailed analysis of neural versus phrase-based statistical machine translation outputs, leveraging high quality post-edits performed by professional translators on the IWSLT data. In this analysis, we focus on two language directions with different characteristics: English–German, known to be particularly hard because of morphology and syntactic differences, and English–French, where PBMT systems typically reach outstanding quality and thus represent a strong competitor for NMT. Our analysis provides useful insights on what linguistic phenomena are best modelled by neural models – such as the reordering of verbs and nouns – while pointing out other aspects that remain to be improved – like the correct translation of proper nouns.

Comparing Language Related Issues for NMT and PBMT between German and English

Abstract This work presents an extensive comparison of language related problems for neural machine translation and phrase-based machine translation between German and English. The explored issues are related both to the language characteristics as well as to the machine translation process and, although related, are going beyond typical translation error classes. It is shown that the main advantage of the NMT system consists of better handling of verbs, English noun collocations, German compound words, phrase structure as well as articles. In addition, it is shown that the main obstacles for the NMT system are prepositions, translation of English (source) ambiguous words and generating English (target) continuous tenses. Although in total there are less issues for the NMT system than for the PBMT system, many of them are complementary – only about one third of the sentences deals with the same issues, and for about 40% of the sentences the issues are completely different. This means that combination/hybridisation of the NMT and PBMT approaches is a promising direction for improving both types of systems.

Using Word Embeddings to Enforce Document-Level Lexical Consistency in Machine Translation

Abstract We integrate new mechanisms in a document-level machine translation decoder to improve the lexical consistency of document translations. First, we develop a document-level feature designed to score the lexical consistency of a translation. This feature, which applies to words that have been translated into different forms within the document, uses word embeddings to measure the adequacy of each word translation given its context. Second, we extend the decoder with a new stochastic mechanism that, at translation time, allows to introduce changes in the translation oriented to improve its lexical consistency. We evaluate our system on English–Spanish document translation, and we conduct automatic and manual assessments of its quality. The automatic evaluation metrics, applied mainly at sentence level, do not reflect significant variations. On the contrary, the manual evaluation shows that the system dealing with lexical consistency is preferred over both a standard sentence-level and a standard document-level phrase-based MT systems.

A Linguistic Evaluation of Rule-Based, Phrase-Based, and Neural MT Engines

AbstractIn this paper, we report an analysis of the strengths and weaknesses of several Machine Translation (MT) engines implementing the three most widely used paradigms. The analysis is based on a manually built test suite that comprises a large range of linguistic phenomena. Two main observations are on the one hand the striking improvement of an commercial online system when turning from a phrase-based to a neural engine and on the other hand that the successful translations of neural MT systems sometimes bear resemblance with the translations of a rule-based MT system.

Filtering Reordering Table Using a Novel Recursive Autoencoder Model for Statistical Machine Translation

In phrase-based machine translation (PBMT) systems, the reordering table and phrase table are very large and redundant. Unlike most previous works which aim to filter phrase table, this paper proposes a novel deep neural network model to prune reordering table. We cast the task as a deep learning problem where we jointly train two models: a generative model to implement rule embedding and a discriminative model to classify rules. The main contribution of this paper is that we optimize the reordering model in PBMT by filtering reordering table using a recursive autoencoder model. To evaluate the performance of the proposed model, we performed it on public corpus to measure its reordering ability. The experimental results show that our approach obtains high improvement in BLEU score with less scale of reordering table on two language pairs: English-Chinese (+0.28) and Uyghur-Chinese (+0.33) MT.

Hierarchical Back-off Modeling of Hiero Grammar based on Non-parametric Bayesian Model

In hierarchical phrase-based machine translation, a rule table is automatically learned by heuristically extracting synchronous rules from a parallel corpus. As a result, spuriously many rules are extracted which may be composed of various incorrect rules. The larger rule table incurs more run time for decoding and may result in lower translation quality. To resolve the problems, we propose a hierarchical back-off model for Hiero grammar, an instance of a synchronous context free grammar (SCFG), on the basis of the hierarchical Pitman-Yor process. The model can extract a compact rule and phrase table without resorting to any heuristics by hierarchically backing off to smaller phrases under SCFG. Inference is efficiently carried out using two-step synchronous parsing of Xiao et al., (2012) combined with slice sampling. In our experiments, the proposed model achieved higher or at least comparable translation quality against a previous Bayesian model on various language pairs; German/French/Spanish/JapaneseEnglish. When compared against heuristic models, our model achieved comparable translation quality on a full size GermanEnglish language pair in Europarl v7 corpus with significantly smaller grammar size; less than 10% of that for heuristic model.