Example-based Machine Translation System Research Articles

Analogical-Based Translation Hypothesis Derivation with Structural Semantics for English to Malay Example-Based Machine Translation

An analogical-based approach towards meaning preservation by transferring the source language meaning structure to the target language in the recombination process of an English to Malay Example-based Machine Translation system is presented. The meaning structure is built on top of the current synchronized translation examples pair representation with the incorporation of a layer of semantic annotation at the structural level. This meaning structure provides a consistent medium allowing the derivation of translation hypothesis using analogical-based approach throughout the automated translation process. The complexity of the structural transformation in the final recombination process is relaxed with this analogical-based derivation approach. The preliminary experiment demonstrates that the English to Malay automated translation is improved using the analogical-based approach. The best evaluation score is obtained using the Bilingual Evaluation Understudy metric, showing improvement of 37.06%.

Meaning preservation in Example-based Machine Translation with structural semantics

The main tasks in Example-based Machine Translation (EBMT) comprise of source text decomposition, following with translation examples matching and selection, and finally adaptation and recombination of the target translation. As the natural language is ambiguous in nature, the preservation of source text’s meaning throughout these processes is complex and challenging. A structural semantics is introduced, as an attempt towards meaning-based approach to improve the EBMT system. The structural semantics is used to support deeper semantic similarity measurement and impose structural constraints in translation examples selection. A semantic compositional structure is derived from the structural semantics of the selected translation examples. This semantic compositional structure serves as a representation structure to preserve the consistency and integrity of the input sentence’s meaning structure throughout the recombination process. In this paper, an English to Malay EBMT system is presented to demonstrate the practical application of this structural semantics. Evaluation of the translation test results shows that the new translation framework based on the structural semantics has outperformed the previous EBMT framework.

Large aligned treebanks for syntax-based machine translation

We present a collection of parallel treebanks that have been automatically aligned on both the terminal and the non-terminal constituent level for use in syntax-based machine translation. We describe how they were constructed and applied to a syntax- and example-based machine translation system called Parse and Corpus-Based Machine Translation (PaCo-MT). For the language pair Dutch to English, we present non-terminal alignment evaluation scores for a variety of tree alignment approaches. Finally, based on the parallel treebanks created by these approaches, we evaluate the MT system itself and compare the scores with those of Moses, a current state-of-the-art statistical MT system, when trained on the same data.

A Pure EBMT Approach for English to Hindi Sentence Translation System

The paper focuses on Example Based Machine Translation (EBMT) system that translates sentences from English to Hindi. It uses the parallel corpus for translating sentences. Development of a machine translation (MT) system typically demands a large volume of computational resources. Requirement of computational resources (for example, rules) is much less in respect of EBMT. This makes development of EBMT systems for English to Hindi translation feasible, where availability of large-scale computational resources is still scarce. Example based machine translation relies on the database for its translation. The frequency of word occurrence is important for translation in EBMT in the following research.

Urdu to English Machine Translation using Bilingual Evaluation Understudy

Machine Translation (MT) is exigent because it involves several thorny subtasks such as intrinsic language ambiguities, linguistic complexities and diversities between source and target language. Usually MT depends upon rules that provide linguistic information. At present, the corpus based MT approaches are used that include techniques like Example Based MT (EBMT) and Statistical MT (SMT). In addition to others, both of these corpus based techniques have different frameworks in the contemporary data-driven paradigm. SMT systems generate outputs using probabilities, whereas EBMT systems translate input text by matching examples from large amount of training data. Urdu MT is in its infancy with very limited availability of required data and computational resources. In this paper, we analyzed and evaluated the main MT techniques using qualitative as well as quantitative approaches. Strengths and weaknesses of each technique have been brought to light through special focus and discussion on examples from Urdu language MT literature. We evaluated the automated machine translated outputs using Bilingual Evaluation Understudy (BLEU). The EBMT approach produced the highest accuracy of 84.21% whereas the accuracy of the online SMT system is 62.68%. We found that BLUE scores of machine translated long Urdu sentences are low in comparison with long sentences. Similarly source text containing low frequency words affect the quality of Urdu machine translation negatively. Experiments and findings section of this paper explicate our reported results in detail. The paper concludes with proposal of future directions for research in Urdu machine translation.

Chinese-Japanese Machine Translation Exploiting Chinese Characters

The Chinese and Japanese languages share Chinese characters. Since the Chinese characters in Japanese originated from ancient China, many common Chinese characters exist between these two languages. Since Chinese characters contain significant semantic information and common Chinese characters share the same meaning in the two languages, they can be quite useful in Chinese-Japanese machine translation (MT). We therefore propose a method for creating a Chinese character mapping table for Japanese, traditional Chinese, and simplified Chinese, with the aim of constructing a complete resource of common Chinese characters. Furthermore, we point out two main problems in Chinese word segmentation for Chinese-Japanese MT, namely, unknown words and word segmentation granularity, and propose an approach exploiting common Chinese characters to solve these problems. We also propose a statistical method for detecting other semantically equivalent Chinese characters other than the common ones and a method for exploiting shared Chinese characters in phrase alignment. Results of the experiments carried out on a state-of-the-art phrase-based statistical MT system and an example-based MT system show that our proposed approaches can improve MT performance significantly, thereby verifying the effectiveness of shared Chinese characters for Chinese-Japanese MT.

The CMU-EBMT machine translation system

This paper presents an in-depth description of the features of the open-source CMU-EBMT example-based machine translation system. CMU-EBMT is a complete end-to-end system including lexicon induction, word and phrase alignment, corpus indexing and lookup, language model, decoder, and parameter tuning components. While it does not require them, it can take advantage of external alignment information and other annotations provided by GIZA++ and other systems. To illustrate a recent addition to CMU-EBMT, experiments are presented which show an improvement of 0.16 BLEU points (0.9% relative) on a cross-validated small-data English---Haitian translation task when using a new set of fine-grained log-linear feature values representing language model match lengths in addition to language model probabilities.

Panning for EBMT gold, or “Remembering not to forget”

A very useful service to the example-based machine translation (EBMT) community was provided by Harold Somers in his summary article which appeared in 1999, and was extended in our 2003 book Recent advances in example-based machine translation. As well as providing a comprehensive review of the paradigm, Somers gives a categorisation of the different instantiations of the basic model. In this paper, we provide a complementary view to that of Somers. Today's EBMT systems learn by analogy. Perhaps even more so than statistical models of translation, one might view these systems as being incapable of forgetting. We researchers and system developers, on the other hand, often forget or are ignorant of techniques and models presented in prior research. The primary aim of this paper is to try to ensure that golden nuggets from past (now quite distantly so) EBMT research papers are gathered together and presented here for a new generation of researchers keen to operate in the paradigm, especially given the spate of recent open-source releases of EBMT systems. We revisit the findings of the previous main research papers, relate them to some of the major research efforts which have taken place since then, and examine especially the prophecies given in the older pieces of work to see the extent to which they have been borne out in the newer research. Given the strong convergence between the leading corpus-based approaches to MT, especially since the introduction of phrase-based statistical MT, a further hope is that these findings may also prove useful to researchers and developers in other areas of MT.

A ranking method for example based machine translation results by learning from user feedback

Example-Based Machine Translation (EBMT) is a corpus based approach to Machine Translation (MT), that utilizes the translation by analogy concept. In our EBMT system, translation templates are extracted automatically from bilingual aligned corpora by substituting the similarities and differences in pairs of translation examples with variables. In the earlier versions of the discussed system, the translation results were solely ranked using confidence factors of the translation templates. In this study, we introduce an improved ranking mechanism that dynamically learns from user feedback. When a user, such as a professional human translator, submits his evaluation of the generated translation results, the system learns “context-dependent co-occurrence rules” from this feedback. The newly learned rules are later consulted, while ranking the results of the subsequent translations. Through successive translation-evaluation cycles, we expect that the output of the ranking mechanism complies better with user expectations, listing the more preferred results in higher ranks. We also present the evaluation of our ranking method which uses the precision values at top results and the BLEU metric.

Extracting and Using Translation Templates in an Example-Based Machine Translation System

A bidirectional English-Filipino machine translation system is developed that extracts translation templates and chunks from a given bilingual English-Filipino corpus. These templates and chunks are then used to translate an input English document to Filipino and vice versa. The system extended the similarity and difference translation template learning algorithms of Cicekli and Guvenir (2003) by refining existing templates and deriving templates from previously learned chunks. Chunk alignment, splitting algorithms, and chunk refinement are also introduced in the training process. Correct extraction of similarity templates and chunks during the learning process led to translation with a low word error rate of 15% for a test document whose sentences exactly match the training set, to a high 86% when the test document is different from the training corpus. Using difference templates alone, the resulting translation has a word error rate of 49% to 85%. Combined use of similarity and difference templates resulted in a low word error rate of 18% when the test document contains sentence patterns matching the training set, to a high 85% when the test document is different from the training corpus. Tests also showed that the translation with the highest score selected from a set of candidate translations is consistently the best choice as validated against automatic evaluation methods.

LEADING BY EXAMPLE: AUTOMATIC TRANSLATION OF SUBTITLES VIA EBMT

This paper describes a project to investigate the scope of the application of Example‐Based Machine Translation (EBMT) to the translation of DVD subtitles and bonus material for English-German and English‐Japanese. The project focused on the development of the EBMT system and its evaluation. This was undertaken as an interdisciplinary study, combining expertise in the areas of multimedia translation, corpus linguistics and natural language processing. The main areas on which this paper focuses are subtitle corpus creation, development of the EBMT system for English‐German, development of evaluation methods for MT output and an assessment of the productiveness of different data types to EBMT.

Inducing translation templates with type constraints

This paper presents a generalization technique that induces translation templates from a given set of translation examples by replacing differing parts in the examples with typed variables. Since the type of each variable is inferred during the learning process, each induced template is also associated with a set of type constraints. The type constraints that are associated with a translation template restrict the usage of the translation template in certain contexts in order to avoid some of the wrong translations. The types of variables are induced using type lattices designed for both the source and target languages. The proposed generalization technique has been implemented as a part of an example-based machine translation system.

A system-theoretical view of EBMT

According to the system theory of von Bertalanffy (1968), Bertalanffy, a "system" is an entity that can be distinguished from its environment and that consists of several parts. System theory investigates the role of the parts, their interaction and the relation of the whole with its environment. System theory of the second order examines how an observer relates to the system. This paper traces some of the recent discussion of example-based machine translation (EBMT) and compares a number of EBMT and statistical MT systems. It is found that translation examples are linguistic systems themselves that consist of words, phrases and other constituents. Two properties of Luhmann's (2002) system theory are discussed in this context: EBMT has focussed on the properties of structures suited for translation and the design of their reentry points, and SMT develops connectivity operators which select the most likely continuations of structures. While technically the SMT and EBMT approaches complement each other, the principal distinguishing characteristic results from different sets of values which SMT and EBMT followers prefer.

Machine Translation Based on Translation Corresponding Tree Structure

A representation schema called translation corresponding tree (TCT) has been applied to a Portuguese to Chinese example-based machine translation system. The translation examples are annotated by the representation of the TCT structure. Each TCT describes not only the syntactic structure of the source sentence (i.e., Portuguese in our system) but also the translation correspondences (i.e., Chinese translation). In addition, the TCT nodes describe the corresponding linguistic relationships between the source and target languages. The translation examples can be effectively represented with this annotation schema and organized in the bilingual knowledge database or example base. In the real machine translation process, the target language is synthesized with higher quality by referring to the TCT translation information.

Example-based machine translation using efficient sentence retrieval based on edit-distance

An Example-Based Machine Translation (EBMT) system, whose translation example unit is a sentence, can produce an accurate and natural translation if translation examples similar enough to an input sentence are retrieved. Such a system, however, suffers from the problem of narrow coverage. To reduce the problem, a large-scale parallel corpus is required and, therefore, an efficient method is needed to retrieve translation examples from a large-scale corpus. The authors propose an efficient retrieval method for a sentence-wise EBMT using edit-distance. The proposed retrieval method efficiently retrieves the most similar sentences using the measure of edit-distance without omissions. The proposed method employs search-space division, word graphs, and an A* search algorithm. The performance of the EBMT was evaluated through Japanese-to-English translation experiments using a bilingual corpus comprising hundreds of thousands of sentences from a travel conversation domain. The EBMT system achieved a high-quality translation ability by using a large corpus and also achieved efficient processing by using the proposed retrieval method.

Comparing example-based and statistical machine translation

In previous work (Gough and Way 2004), we showed that our Example-Based Machine Translation (EBMT) system improved with respect to both coverage and quality when seeded with increasing amounts of training data, so that it significantly outperformed the on-line MT system Logomedia according to a wide variety of automatic evaluation metrics. While it is perhaps unsurprising that system performance is correlated with the amount of training data, we address in this paper the question of whether a large-scale, robust EBMT system such as ours can outperform a Statistical Machine Translation (SMT) system. We obtained a large English-French translation memory from Sun Microsystems from which we randomly extracted a near 4K test set. The remaining data was split into three training sets, of roughly 50K, 100K and 200K sentence-pairs in order to measure the effect of increasing the size of the training data on the performance of the two systems. Our main observation is that contrary to perceived wisdom in the field, there appears to be little substance to the claim that SMT systems are guaranteed to outperform EBMT systems when confronted with ‘enough’ training data. Our tests on a 4.8 million word bitext indicate that while SMT appears to outperform our system for French-English on a number of metrics, for English-French, on all but one automatic evaluation metric, the performance of our EBMT system is superior to the baseline SMT model.

Controlled Translation in an Example-based Environment: What do Automatic Evaluation Metrics Tell Us?

This paper presents an extended, harmonised account of our previous work on integrating controlled language data in an Example-based Machine Translation system. Gough and Way in MT Summit pp. 133---140 (2003) focused on controlling the output text in a novel manner, while Gough and Way (9th Workshop of the EAMT, (2004a), pp. 73---81) sought to constrain the input strings according to controlled language specifications. Our original sub-sentential alignment algorithm could deal only with 1:1 matches, but subsequent refinements enabled n:m alignments to be captured. A direct consequence was that we were able to populate the system's databases with more than six times as many potentially useful fragments. Together with two simple novel improvements --- correcting a small number of mistranslations in the lexicon, and allowing multiple translations in the lexicon --- translation quality improves considerably. We provide detailed automatic and human evaluations of a number of experiments carried out to test the quality of the system. We observe that our system outperforms the rule-based on-line system Logomedia on a range of automatic evaluation metrics, and that the `best' translation candidate is consistently highly ranked by our system. Finally, we note in a number of tests that the BLEU metric gives objectively different results than other automatic evaluation metrics and a manual evaluation. Despite these conflicting results, we observe a preference for controlling the source data rather than the target translations.

WEBMT: Developing and Validating an Example-Based Machine Translation System Using the World Wide Web

We have developed an example-based machine translation (EBMT) system that uses the World Wide Web for two different purposes: First, we populate the system's memory with translations gathered from rule-based MT systems located on the Web. The source strings input to these systems were extracted automatically from an extremely small subset of the rule types in the Penn-II Treebank. In subsequent stages, the source, target translation pairs obtained are automatically transformed into a series of resources that render the translation process more successful. Despite the fact that the output from on-line MT systems is often faulty, we demonstrate in a number of experiments that when used to seed the memories of an EBMT system, they can in fact prove useful in generating translations of high quality in a robust fashion. In addition, we demonstrate the relative gain of EBMT in comparison to on-line systems. Second, despite the perception that the documents available on the Web are of questionable quality, we demonstrate in contrast that such resources are extremely useful in automatically postediting translation candidates proposed by our system.