Translation Patterns, Linguistic Knowledge and Complexity in an Approach to EBMT

McTait, Kevin

doi:10.1007/978-94-010-0181-6_11

Cited by 9 publications

(19 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Some EBMT systems use a bilingual corpus to find translations of the parts of a given sentence, and combine these partial solutions to get the translation of the whole sentence. On the other hand, some other EBMT systems (Kaji et al, 1992;Cicekli and Güvenir, 2001;Brown, 2003;Carl, 2003;Cicekli and Güvenir, 2003;McTait, 2003) extract translation templates from example sentences in a given bilingual corpus and use these translation templates in the translation of other sentences. The main differences between these EBMT systems are the assumptions made on the structure of the bilingual corpus and their generalization techniques.…”

Section: Introductionmentioning

confidence: 99%

Inducing translation templates with type constraints

Çiçekli

2007

Machine Translation

View full text Add to dashboard Cite

Abstract. 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 language and the target language. The proposed generalization technique has been implemented as a part of an EBMT system.

show abstract

Section: Introductionmentioning

confidence: 99%

Inducing translation templates with type constraints

Çiçekli

2007

Machine Translation

View full text Add to dashboard Cite

show abstract

“…However, it gradually becomes evident that some amount of linguistic knowledge is necessary (see [12] for the case of SMT). On the other hand, all these methods crucially depend on large bitexts [8]. But bitexts are rare [1] and, quite often, of questionable linguistic quality.…”

Section: Main Featuresmentioning

confidence: 99%

“…In this context, proposed ways for fragmenting sentences range from the exploitation of highly structured representations of linguistic knowledge [14] to the establishment of string correspondences accompanied by little/trivial linguistic knowledge representation [8]. However, methods combining sub-sentential strings face the problem of boundary friction.…”

Section: Main Featuresmentioning

confidence: 99%

Pattern Matching-Based System for Machine Translation (MT)

Tambouratzis

Sofianopoulos

Spilioti

et al. 2006

Advances in Artificial Intelligence

View full text Add to dashboard Cite

Abstract. The innovative feature of the system presented in this paper is the use of pattern-matching techniques to retrieve translations resulting in a flexible, language-independent approach, which employs a limited amount of explicit a priori linguistic knowledge. Furthermore, while all state-of-the-art corpus-based approaches to Machine Translation (MT) rely on bitexts, this system relies on extensive target language monolingual corpora. The translation process distinguishes three phases: 1) pre-processing with 'light' rule and statisticsbased NLP techniques 2) search & retrieval, 3) synthesising. At Phase 1, the source language sentence is mapped onto a lemma-to-lemma translated string. This string then forms the input to the search algorithm, which retrieves similar sentences from the corpus (Phase 2). This retrieval process is performed iteratively at increasing levels of detail, until the best match is detected. The best retrieved sentence is sent to the synthesising algorithm (Phase 3), which handles phenomena such as agreement.

show abstract

“…Matsumoto and Kitamura (1995) derive generalised examples by means of semantic features. Brown (1999Brown ( , 2000Brown ( , 2003 automatically generates equivalence classes using word-level clustering techniques, while Güvenir et al (Güvenir and Tunç, 1996;Güvenir and Cicekli, 1998;Cicekli and Güvenir, 2003) for Turkish-English and McTait and Trujillo (1999) and McTait (2003) for English-Spanish implement transfer-rule induction from bitexts. Perhaps the most similar technique to ours is that of Block (2000), who replaces certain words by part-of-speech tags rather than marker tags, as here.…”

Section: Marker-based Ebmtmentioning

confidence: 99%

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

Way

Gough

2005

Mach Translat

View full text Add to dashboard Cite

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.

show abstract

Translation Patterns, Linguistic Knowledge and Complexity in an Approach to EBMT

Cited by 9 publications

References 10 publications

Inducing translation templates with type constraints

Inducing translation templates with type constraints

Pattern Matching-Based System for Machine Translation (MT)

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

Contact Info

Product

Resources

About