Steps Toward Knowledge-Based Machine Translation

Carbonell, Jaime G.; Cullingford, Richard E.; Gershman, Anatole

doi:10.1109/tpami.1981.4767124

Cited by 47 publications

(30 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The fundamental reason is that, if a formerly translated sentence happens once more, the same translation is liable to be right once more. [6].…”

Section: Example Based Machine Translationmentioning

confidence: 99%

Machine Translation Techniques and their Comparative Study

Ashraf¹,

Ahmad²

2015

IJCA

View full text Add to dashboard Cite

Human beings exhibit a striking quality of communicating with each other. Communication by means of a system of communication based upon words and the combination of words into sentences, referred as linguistic communication.None of non-human species have such a system of communication in place that's comparable to human language. What makes languages of human varied and different, are features of duality and arbitrary. In annuals of Anthropology, language is considered as a primary tool for studying the culture of a civilization, what we speak influences what we think, what we feel and what we believe. Culture is transmitted through language. Humans learn their culture through language. Its inquisitive nature of human and passion to travel across the world, warrants different cultures interact with each other, the means to achieve this is through human language, often interacting cultures communicate through different languages. As such, it's essential that humans translate and interpret languages of different cultures for understand their rituals, business and allied activities. With advancements in technology, computer systems have facilitated the translations of languages and achieved results in minimal amount of time, though these systems do not produce exact translated verse but enough and relevant information that could be used by the information professionals to understand the nature of information contained in the document, tools like Babelfish and Google Translator are examples of such systems. Numerous techniques have been developed to automate the translation process and these are termed under Machine Translation, which can be defined as a task of automatically converting one natural language into another, preserving the meaning of the input text, and producing fluent text in the output language. These automated translation systems use state of art technology with wideranging dictionaries and a collection of linguistic rules that translate one language into another without relying on human translators. The motivation of this research is to have a comparative study of machine translation techniques used for multilingual translation vis-à-vis efficiency, ease of use, space-time complexity and creation of experimental framework for comparing machine translation techniques using open-source translation tools.

show abstract

“…The fundamental reason is that, if a formerly translated sentence happens once more, the same translation is liable to be right once more. [6].…”

Section: Example Based Machine Translationmentioning

confidence: 99%

Machine Translation Techniques and their Comparative Study

Ashraf¹,

Ahmad²

2015

IJCA

View full text Add to dashboard Cite

show abstract

“…Also, methods for language-specific sentence-level restructuring transformations are applied after the translation has been obtained. Finally, in [22] a very large amount of context of the target language is used to find and rank N-grams containing the phrase translations of the source text. However, to the best of our knowledge, there are no works that systematically evaluate their techniques using very short phrases only, like those in business process models.…”

Section: Related Workmentioning

confidence: 99%

Automatic Business Process Model Translation with BPMT

Batoulis¹,

Eid-Sabbagh²,

Leopold³

et al. 2013

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

View full text Add to dashboard Cite

Summary. Nowadays, many enterprises use business process models for documenting and supporting their operations. As many enterprises have branches in several countries and provide similar services throughout the globe, there is high potential for re-using these process models. However, the language barrier is a major obstacle for the successful re-use of process models, especially in multi-national companies. In this paper, we address this problem by presenting the Business Process Model Translator (BPMT), a technique for the automated translation of business process models that eases the re-use of business process models and reduces redundant work in multi-national companies. It builds upon the state-of-the-art machine translation system Moses and extends it with word and translation disambiguation considering the context of the domain. As a result, the BPMT can successfully deal with the compact and special language fragments that are typically found in business process models. A two-fold evaluation with the BLEU metric and an expert survey showed improvements of our approach over Moses.

show abstract

“…The output of analysis is a deep syntactic dependency that normalizes over syntactic phenomena such as passivization and morphological expressions of tense, number, 8 Another corpus-based MT approach that relies heavily on TL language modeling is context-based MT (CBMT) (Carbonell et al 2006), which does not require the presence of parallel corpora, although it needs a very large surface-based dictionary. 9 Another non-interlingual non-transfer approach is Shake-and-Bake MT (Beaven 1992;Whitelock 1992) which overgenerates TL sentences and symbolically constrains the output by "parsing it."…”

Section: Handling Translation Divergencesmentioning

confidence: 99%

Symbolic-to-statistical hybridization: extending generation-heavy machine translation

Habash

Dorr

Monz

2009

Machine Translation

View full text Add to dashboard Cite

The last few years have witnessed an increasing interest in hybridizing surface-based statistical approaches and rule-based symbolic approaches to machine translation (MT). Much of that work is focused on extending statistical MT systems with symbolic knowledge and components. In the brand of hybridization discussed here, we go in the opposite direction: adding statistical bilingual components to a symbolic system. Our base system is Generation-heavy machine translation (GHMT), a primarily symbolic asymmetrical approach that addresses the issue of Interlingual MT resource poverty in source-poor/target-rich language pairs by exploiting symbolic and statistical target-language resources. GHMT's statistical components are limited to target-language models, which arguably makes it a simple form of a hybrid system. We extend the hybrid nature of GHMT by adding statistical bilingual components. We also describe the details of retargeting it to Arabic-English MT. The morphological richness of Arabic brings several challenges to the hybridization task. We conduct an extensive evaluation of multiple system variants. Our evaluation shows that this new variant of GHMT-a primarily symbolic system extended with monolingual and bilingual statistical components-has a higher degree of grammaticality than a phrase-based statistical MT system, where grammaticality is measured in terms of correct verb-argument realization and long-distance dependency translation.

show abstract

Steps Toward Knowledge-Based Machine Translation

Cited by 47 publications

References 11 publications

Machine Translation Techniques and their Comparative Study

Machine Translation Techniques and their Comparative Study

Automatic Business Process Model Translation with BPMT

Symbolic-to-statistical hybridization: extending generation-heavy machine translation

Contact Info

Product

Resources

About