Statistical Machine Translation System Research Articles

Word Sense Disambiguation applied to Assamese-Hindi Bilingual Statistical Machine Translation

Word Sense Disambiguation (WSD) is concerned with automatically assigning the appropriate sense to an ambiguous word. WSD is an important task and plays a crucial role in many Natural Language Processing (NLP) applications. A Statistical Machine Translation (SMT) system translates a source into a target language based on phrase-based statistical translation. MT plays a crucial role in a WSD system, as a source language word may be associated with multiple translations in the target language. This study aims to apply WSD to the input of the MT system to enhance the disambiguation output. Hindi WordNet was used by selecting the most frequent synonym to obtain the most accurate translation. This study also compared Naïve Bayes (NB) and Decision Tree (DT) to test and build a WSD model. NB was more appropriate for the WSD task than DT when evaluated in the Weka machine learning toolkit. To the best of our knowledge, no such work has been carried out yet for the Assamese Indo-Aryan language. The applied WSD achieved better results than the baseline MT system without embedding the WSD module. The results were analyzed by linguist scholars. Furthermore, the Assamese-Hindi transliteration system was merged with the baseline MT system for the translation of proper nouns. This study marks a remarkable contribution to Assamese NLP, which is a low computationally aware Indian language.

An empirical analysis on statistical and neural machine translation system for English to Mizo language

An empirical analysis on statistical and neural machine translation system for English to Mizo language

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.

Automatic Resource Augmentation for Machine Translation in Low Resource Language: EnIndic Corpus

Parallel corpus is the primary ingredient of machine translation. It is required to train the statistical machine translation (SMT) and neural machine translation (NMT) systems. There is a lack of good quality parallel corpus for Hindi to English. Comparable corpora for a given language pair are comparatively easy to find, but this cannot be used directly in SMT or NMT systems. As a result, we generate a parallel corpus from the comparable corpus. For this purpose, the sentences (which are translations of each other) are mined from the comparable corpus to prepare the parallel corpus. The proposed algorithm uses the length of the sentence and word translation model to align sentence pairs that are translations of each other. Then, the sentence pairs that are poor translations of each other (measured by a similarity score based on IBM model 1 translation probability) are filtered out. We apply this algorithm to comparable corpora, which are crawled from speeches of the President and Vice-President of India, and mined parallel corpora out of them. The prepared parallel corpus contains good quality aligned sentences (with 96.338% f-score). Subsequently, incorrect sentence pairs are filtered out manually to make the corpus in qualitative practical use. Finally, we gather various sentences from different sources to prepare the EnIndic corpus, which comprises 1,656,207 English-Hindi sentence pairs (miscellaneous domain). We have deployed this prepared largest English-Hindi parallel corpus at https://github.com/debajyoty/EnIndic.git and the source code at https://github.com/debajyoty/EnIndicSourceCode.git.

Fuzzy Influenced Process to Generate Comparable to Parallel Corpora

Data-driven supervised approaches rely on the parallel corpus. Due to lack of data and resources availability, it has become more difficult to achieve accurate outputs. In addition, the efficiency of the machine translation system depends on the quality of the used corpora. Hindi still lacks good quality parallel corpora and needs more resources for accurate machine translation. Comparable corpora are easily available compared to parallel corpora, but they cannot be used directly in machine translation. In our present research, we propose an algorithm to mine these comparable corpora from the web, and generate the parallel corpora automatically. Machine translation systems, system combination approach, and IR-based technique join their hands together to choose the set of sentence pairs. Then the sentence pairs having the best score are chosen to prepare the final parallel corpora. The primary modules of this architecture are fuzzy logic-based evaluation metric, information retrieval module, statistical machine translation system, Google neural machine translation system, Microsoft machine translation system, and system combination module for machine translation. For case study, we prepare the Hindi-English parallel corpora of (30825 + 51235) = 82060 sentence pairs. Evaluation results show that the F-Score measurement varies from 95.73 to 96.98 for various data sets. The source code and prepared dataset (comparable and parallel corpus) can be found at https://github.com/debajyoty/Comparable-partallel-Algo2.git.

Reduction of Neural Machine Translation Failures by Incorporating Statistical Machine Translation

This paper proposes a hybrid machine translation (HMT) system that improves the quality of neural machine translation (NMT) by incorporating statistical machine translation (SMT). Therefore, two NMT systems and two SMT systems were built for the Slovenian–English language pair, each for translation in one direction. We used a multilingual language model to embed the source sentence and translations into the same vector space. From each vector, we extracted features based on the distances and similarities calculated between the source sentence and the NMT translation, and between the source sentence and the SMT translation. To select the best possible translation, we used several well-known classifiers to predict which translation system generated a better translation of the source sentence. The proposed method of combining SMT and NMT in the hybrid system is novel. Our framework is language-independent and can be applied to other languages supported by the multilingual language model. Our experiment involved empirical applications. We compared the performance of the classifiers, and the results demonstrate that our proposed HMT system achieved notable improvements in the BLEU score, with an increase of 1.5 points and 10.9 points for both translation directions, respectively.

San-Eng: Sanskrit to English Translator using Machine Learning

In the past decade, machine learning has made great strides in improving automatic translation. In general, machine learning algorithms have been able to achieve better translation quality by learning from large amounts of data. One approach to machine translation is to use a statistical machine translation (SMT) system. SMT systems learn to translate by statistical methods, using a large parallel corpus of text in multiple languages. The most successful SMT systems are based on the translation model known as the phrase-based translation model. In recent years, a new approach to machine translation has emerged, known as neural machine translation (NMT). NMT systems use artificial neural networks to learn to translate. NMT systems have shown to be very successful in translating between languages that are closely related, such as English and Hindi or Reverse is also true. In recent years, a new approach to machine translation has emerged, known as neural machine translation (NMT). NMT systems use artificial neural networks to learn to translate. NMT systems have shown to be very successful in translating between languages that are closely related, such as English and Hindi or Reverse is also true

Reordering of Source Side for a Factored English to Manipuri SMT System

Similar languages with massive parallel corpora are readily implemented by large-scale systems using either Statistical Machine Translation (SMT) or Neural Machine Translation (NMT). Translations involving low-resource language pairs with linguistic divergence have always been a challenge. We consider one such pair, English-Manipuri, which shows linguistic divergence and belongs to the low resource category. For such language pairs, SMT gets better acclamation than NMT. However, SMT’s more prominent phrase- based model uses groupings of surface word forms treated as phrases for translation. Therefore, without any linguistic knowledge, it fails to learn a proper mapping between the source and target language symbols. Our model adopts a factored model of SMT (FSMT3*) with a part-of-speech (POS) tag as a factor to incorporate linguistic information about the languages followed by hand-coded reordering. The reordering of source sentences makes them similar to the target language allowing better mapping between source and target symbols. The reordering also converts long-distance reordering problems to monotone reordering that SMT models can better handle, thereby reducing the load during decoding time. Additionally, we discover that adding a POS feature data enhances the system’s precision. Experimental results using automatic evaluation metrics show that our model improved over phrase-based and other factored models using the lexicalised Moses reordering options. Our FSMT3* model shows an increase in the automatic scores of translation result over the factored model with lexicalised phrase reordering (FSMT2) by an amount of 11.05% (Bilingual Evaluation Understudy), 5.46% (F1), 9.35% (Precision), and 2.56% (Recall), respectively.

Hyper-parameter optimization in neural-based translation systems: A case study

Abstract Machine translation (MT) is an important use case in natural language processing (NLP) that converts a source language to a target language automatically. Modern intelligent system or artificial intelligence (AI) uses a machine learning approach and the machine has acquired learning ability using datasets. Nowadays, in the MT domain, the neural machine translation (NMT) system has almost replaced the statistical machine translation (SMT) system. The NMT systems use a deep learning framework in their implementation. To achieve higher accuracy during the training of the NMT model, extensive hyper-parameter tuning is required. The paper highlights the significance of hyper-parameter tuning in various machine learning algorithms. And as a case study, in-house experimentation was conducted on a low-resource English–Bangla language pair by designing an NMT system and the significance of various hyper-parameter optimizations was analyzed while evaluating its performance with an automatic metric BLEU. The BLEU scores obtained for the first, second, and third randomly picked test sentences are 4.1, 3.2, and 3.01, respectively.

Translation Quality Regarding Low-Resource, Custom Machine Translations: A Fine-Grained Comparative Study on Turkish-to-English Statistical and Neural Machine Translation Systems

Corpus-based machine translation (MT) has been the main approach to developing and implementing MT systems in both academia and the industry over the last three decades. In this field, the type and size of the corpus used for training MT engines have presented problems for both statistical MT (SMT) systems as well as neural MT (NMT) systems, being the two dominant corpusbased approaches. Moreover, language pairs such as Turkish-English have been understudied within this framework. This article aims to evaluate the translation quality in Turkish-to-English custom MT systems that have been trained on different corpus sizes and types. Two NMT engines and two SMT engines were trained on the KantanMT platform using two different training corpus types with either only domain-specific cardiology corpus or this corpus plus a mixed-domain corpus. The study conducted both automatic evaluations with metrics including BLEU, F-Measure and TER, as well as a comprehensive human evaluation with metrics including fluency, A/B test, and adequacy. Lastly, the study realized a separate, subjective terminology evaluation in order to investigate how differently MT systems handle terminology, as this is a crucial aspect for specific-domain text types such as cardiology. While the automatic evaluation results suggest the SMT engines to perform better than NMT engines, all human evaluators rated the mixed-domain NMT engine as the highest performing one. However, the terminology evaluation task demonstrated SMT to still be able to perform better and to commit less terminology errors, despite the industry and academia shifting toward NMT engines.

An empirical study on English-Mizo Statistical Machine Translation with Bible Corpus

Machine Translation (MT) is the process of automatically converting the text or speech in one natural language to another language with the help of a machine. This work presents a Bidirectional Statistical Machine Translation (SMT) system of an extremely low resource language pair Mizo-English, built in a low resource setting. A total of 30800 sentences are collected from the English Bible dataset and manually translated to Mizo by a native linguistic expert to generate the English-Mizo parallel dataset. After subjecting to various pre-processing steps, the parallel dataset is used to build our MT system using MOSES tools. Our framework uses different tools, such as GIZA++ for creating the Translation Model (TM) and IRSTLM to determine the probability of the target model. The quality of our MT system is evaluated using two automatic evaluation metrics: BLEU and METEOR. Our MT systems are also manually evaluated using two parameters: adequacy and fluency.

Traduction d’unités polylexicales du portugais en français par MT@EC et eTranslation

Translation of Multi-Word Units from Portuguese into French by MT@EC and eTranslation
 This paper aims to determine the extent to which the shift from statistical machine translation (SMT) to neural machine translation (NMT) improved the performance of European Union machine translation systems between 2015 and 2021 in terms of multi-word unit translation and domain coverage. To do so, we chose to test these systems on machine translation into French of multi-word units expressing quantitative and qualitative progression in Portuguese from Portugal. These units consist of the 2-gram ‘cada vez’ and a comparative adjective or adverb (cada vez COMP), and their word-for-word translation into French is not idiomatic (*chaque fois COMP). The most frequent translation into French is ‘de COMP en COMP’. This implies that these multi-word units must be translated ‘en bloc’, but their identification is not straightforward. On the one hand, COMP is not fixed and may include one (mais / plus, menos / moins, maior / plus grand, menor / plus petit, melhor / meilleur – mieux, pior / pire – plus mal) or several words (mais or menos N, ADJ, ADV). On the other hand, the 2-gram ‘cada vez’ can be part of other multi-word units expressing iteration (de cada vez (que)/(à) chaque fois (que)), or ‘dropper’ ([a certain quantity] de cada vez/ à la fois), This raises the challenge of ambiguity, well known to biotranslators and still often problematic for NMT. Moreover, units expressing quantitative or qualitative progression may raise other translation challenges when they are coordinate (with or without repetition of the 2-gram ‘cada vez’), when they are split (cada vez (…) COMP), or when they combine with verbs or nouns to form extended translation units whose translation into French can result in a more concise solution we refer to as ‘lexicalisation’. We established a biotranslation model based on a manually aligned French-Portuguese parallel literary corpus and online searchable French-Portuguese aligned corpora (translation memories). We selected a sample of occurrences of these multi-word units including several translation challenges. These occurrences were selected from a Portuguese journalistic corpus. They belong therefore to general language, whereas the EU's translation memories cover the domains dealt with by its institutions, which represents an additional challenge, considering the critical importance of domain coverage in the data to NMT performance quality. The selected occurrences were translated into French by the EU SMT system in 2015 (MT@EC) and 2019 (eTranslation Legacy) and by eTranslation (the EU NMT system) in 2019 and 2021. Firstly, MT output was analysed according to two general criteria: ‘non- literality’, that is translation into French without ‘chaque’, and acceptability from a semantic point of view, that is MT output without any false meaning, opposite meaning or nonsense. Then we looked at specific challenges, some of which could lead to original solutions, worthy of a professional human translator, such as lexicalisation, change of grammatical category or ‘recategorisation’ and ‘naturalisation’, that is phraseological or syntactic rearrangement that makes the target text more idiomatic. The results show that MT is improving, especially according to the criterion of non-literality. Original solutions are still rare, but they are diversifying in NMT output. Nevertheless, NMT remains imperfect, not least because of the inherent ambiguity of natural languages and the inevitable gaps in the data on which these systems are based. The results also demonstrate the importance of human intervention in the maintenance of the systems learning automatically, since the quality of SMT system’s output decreases between 2015 and 2019, when all efforts were focused on improving EU NMT system. Finally, results reveal the dangers of using English as a pivot language when translating from one Romance language into another, and the need to train future translators in NMT and post-editing.

Unsupervised SMT: an analysis of Indic languages and a low resource language

ABSTRACT The rapid globalisation in language technology and the Internet’s fast expansion have brought nations and their cultures close together, and the demand for inter-language interactions has risen enormously. However, in many low-resource languages (LRL) pairings and areas, Machine Translation (MT) is still not viable because of a lack of parallel data. The challenge of MT is still unsolved. Recent studies employing monolingual datasets have shown excellent outcomes in Phrase-based Statistical MT (PBSMT) and Neural MT (NMT) systems. However, earlier researchers have demonstrated that unsupervised Statistical MT surpasses unsupervised NMT, especially for different language pairings. The study unveils the compendium of ten unsupervised SMT systems translation tasks utilizing a monolingual dataset from the Dravidian and Indo-Aryan language families; and a low-resource endangered language. The machine-translated experimental outcomes examined the system using different tokenizers and investigated them for various language pairs using different evaluation metrics for various iterations. The statistical significance of test results has been computed for each evaluation metric to check the true system quality of the translation tasks.

Improving neural machine translation with POS-tag features for low-resource language pairs

Integrating linguistic features has been widely utilized in statistical machine translation (SMT) systems, resulting in improved translation quality. However, for low-resource languages such as Thai and Myanmar, the integration of linguistic features in neural machine translation (NMT) systems has yet to be implemented. In this study, we propose transformer-based NMT models (transformer, multi-source transformer, and shared-multi-source transformer models) using linguistic features for two-way translation of Thai-to-Myanmar, Myanmar-to-English, and Thai-to-English. Linguistic features such as part-of-speech (POS) tags or universal part-of-speech (UPOS) tags are added to each word on either the source or target side, or both the source and target sides, and the proposed models are conducted. The multi-source transformer and shared-multi-source transformer models take two inputs (i.e., string data and string data with POS tags) and produce string data or string data with POS tags. A transformer model that utilizes only word vectors was used as the first baseline model for comparison with the proposed models. The second baseline model, an Edit-Based Transformer with Repositioning (EDITOR) model, was also used to compare with our proposed models in addition to the baseline transformer model. The findings of the experiments show that adding linguistic features to the transformer-based models enhances the performance of a neural machine translation in low-resource language pairs. Moreover, the best translation results were yielded using shared-multi-source transformer models with linguistic features resulting in more significant Bilingual Evaluation Understudy (BLEU) scores and character n-gram F-score (chrF) scores than the baseline transformer and EDITOR models.

Review on Machine Translation from English to Kannada

Abstract: Interlingual is a machine translation tool that uses anartificial language to convey the meaning of reallanguages. The process of converting text from one language to another is known as machine translation.This study provides a better model of machine translation system for English-to-Kannada sentence translation that employs statistically based techniques. Here, we use Moses approach. Moses is a statistical machine translation system. systems are trained on huge amounts of parallel data as well as even bigger amounts of monolingual data in statistical machine translation. Parallel data is a set of sentences in two languages that are sentence- aligned, meaning that each sentence in one languageis matched with its translated counterpart in the other. Moses training technique takes the parallel data and infers translation correspondences betweenthe two languages of interest by looking for co- occurrences of words and segments. The two main components in Moses are the training pipeline and the decoder. The training pipeline consists of a set oftools that take raw data and convert it into a machinetranslation model. The Moses decoder determines thehighest scoring sentence in the target language that matches a given source sentence.

Improved Unsupervised Statistical Machine Translation via Unsupervised Word Sense Disambiguation for a Low-Resource and Indic Languages

ABSTRACT Besides word order, word choice is a key stumbling block for machine translation (MT) in morphologically rich languages due to homonyms and polysemous difficulties. On the other hand, un-translated/improperly translated words are a severe issue for Statistical Machine Translation (SMT) models. The quantity of parallel training corpus has limited unsupervised SMT (USMT) systems. Still, current research lines have successfully trained SMT systems in an unsupervised manner using monolingual data alone. However, there is still a need to enhance the translation quality of the MT output due to unaligned and improperly sensed words. This problem is addressed by incorporating unsupervised Word Sense Disambiguation (WSD) into the decoding phase of USMT. The work provided a compendium of SMT systems for five translation tasks, i.e. En→Indic languages for the WMT test dataset and evaluated on BLEU and METEOR evaluation metrics. The studies were performed on En→Hi, En→Kn, En→Ta, En→Te, and En→Be tasks and showed an improvement in BLEU points by 2.3, 2.68, 0.78, 2.32, and 1.79, respectively, and METEOR points by 1.07, 1.34, 0.72, 0.693, and 1.191, respectively, over the baseline model.

Knowledge Distillation: A Method for Making Neural Machine Translation More Efficient

Neural machine translation (NMT) systems have greatly improved the quality available from machine translation (MT) compared to statistical machine translation (SMT) systems. However, these state-of-the-art NMT models need much more computing power and data than SMT models, a requirement that is unsustainable in the long run and of very limited benefit in low-resource scenarios. To some extent, model compression—more specifically state-of-the-art knowledge distillation techniques—can remedy this. In this work, we investigate knowledge distillation on a simulated low-resource German-to-English translation task. We show that sequence-level knowledge distillation can be used to train small student models on knowledge distilled from large teacher models. Part of this work examines the influence of hyperparameter tuning on model performance when lowering the number of Transformer heads or limiting the vocabulary size. Interestingly, the accuracy of these student models is higher than that of the teachers in some cases even though the student model training times are shorter in some cases. In a novel contribution, we demonstrate for a specific MT service provider that in the post-deployment phase, distilled student models can reduce emissions, as well as cost purely in monetary terms, by almost 50%.

Improving Word Alignment by Adding Gromov-Wasserstein into Attention Neural Network

Statistical machine translation systems usually break the translation task into two or more subtasks and an important one is finding word alignments over a parallel sentence bilingual corpus. We address the problem of introducing word alignment for language pairs by developing a novel neural network model that can applied to other generative alignment models. We use Multi-layer attention model and multi-layer model with multi-head-attention mechanism on each layer provides superior translation quality. It can be trained on bilingual data without relying on word alignment. In this paper, we cast the correspondence problem directly as an optimal distance problem. We use the Gromov-Wasserstein distance to calculated how similarities between word pairs are related across languages. The resulting alignments dramatically outperform the GIZA++ and FastAlign approach, these alignments are comparable on public data sets.

Translational equivalence in Statistical Machine Translation or meaning as co-occurrence

In this paper, we will describe the current state-of-the-art of Statistical Machine Translation (SMT), and reflect on how SMT handles meaning. Statistical Machine Translation is a corpus-based approach to MT: it de-rives the required knowledge to generate new translations from corpora. General-purpose SMT systems do not use any formal semantic representa-tion. Instead, they directly extract translationally equivalent words or word sequences – expressions with the same meaning – from bilingual parallel corpora. All statistical translation models are based on the idea of word alignment, i.e., the automatic linking of corresponding words in parallel texts. The first generation SMT systems were word-based. From a linguistic point of view, the major problem with word-based systems is that the mean-ing of a word is often ambiguous, and is determined by its context. Current state-of-the-art SMT-systems try to capture the local contextual dependen-cies by using phrases instead of words as units of translation. In order to solve more complex ambiguity problems (where a broader text scope or even domain information is needed), a Word Sense Disambiguation (WSD) module is integrated in the Machine Translation environment.

Preventing translation quality deterioration caused by beam search decoding in neural machine translation using statistical machine translation

Decoding is an important part of machine translation systems, and the most popular inference algorithm used here is beam search. Beam search algorithm improves translation by allowing a larger search space to be traversed than greedy search. However, as the beam width increases, the translation performance declines after a certain point in neural machine translation (NMT). This problem is usually not observed in statistical machine translation (SMT) due to the decoding method. This paper proposes a hybrid system-based method that uses SMT predictions to prevent quality deterioration in the beam search algorithm used in NMT decoding. Our approach is based on the reranking n-best list of NMT according to the SMT system translation sentence. We propose two different algorithms for reranking NMT n-best lists. The first algorithm uses the length information of the SMT outputs. In contrast, the second uses a word-based similarity approach with the Jaccard Index, the Dice’s Coefficient, and the Overlap Coefficient. Experiments on three different language pairs show that the method we propose prevents the decrease in translation quality and produces a gain of 1.3 BLEU and 1.6 METEOR for different beam sizes and 1.8 BLEU and 2.1 METEOR average scores compared to the baseline results.