Words In Source Language Research Articles

The Application of Translation Strategies to Cope With Equivalence Problems in Translating Texts

Translation is a process to render the meaning from the source text into the target text. A translator, however, will find some problems during translation process. Equivalence is the case which often appears (i.e. culture specific concept, the source-language concept is not lexicalized in the target language, source-language word is semantically complex, etc). To cope with equivalnce problems in translation process, some experts suggest some strategies which can be applied in doing translation. Some strategies are transference, naturalization, cultural equivalent, etc. The strategies which often appears in the example texts in this paper are transference, naturalization, descriptive equivalent, couplet and through-translation. It is recomended that translator apply the strategies if only there is no equivalence problem in target language.

Resource creation and development of an English-Bangla back transliteration system

In this paper we present the development of an English-Bangla transliteration parallel corpus and used it to develop and evaluate some of the popular computational models to transliterate Bangla texts written in Romanized English, back to its original script. Accordingly, we have developed differen t techniques to generate an English-Bangla parallel transliterated lexicon of around 100,000 words. The proposed lexicon of English-Bangla transliterated word pairs along with some of the language specific orthographic as well as phonetic information rules are used to develop two different computational models namely, the joint source channel model and the phrase based SMT model, to automatically identify, extract and learn the transliteration unit (TU) pairs from both the source and target language words. Both the models are used to predict the top 5 best possible outcome of the given input text. Both the models have been evaluated with a set of 20000 Romanized transliterated Bangla test words. Our initial evaluation results clearly shows that performance of the SMT model slightly surpasses the performance joint source channel model.

Word segmentation and pronunciation extraction from phoneme sequences through cross-lingual word-to-phoneme alignment

In this paper, we study methods to discover words and extract their pronunciations from audio data for non-written and under-resourced languages. We examine the potential and the challenges of pronunciation extraction from phoneme sequences through cross-lingual word-to-phoneme alignment. In our scenario a human translator produces utterances in the (non-written) target language from prompts in a resource-rich source language. We add the resource-rich source language prompts to help the word discovery and pronunciation extraction process. By aligning the source language words to the target language phonemes, we segment the phoneme sequences into word-like chunks. The resulting chunks are interpreted as putative word pronunciations but are very prone to alignment and phoneme recognition errors. Thus we suggest our alignment model Model 3P that is particularly designed for cross-lingual word-to-phoneme alignment. We present two different methods (source word dependent and independent clustering) that extract word pronunciations from word-to-phoneme alignments and compare them. We show that both methods compensate for phoneme recognition and alignment errors. We also extract a parallel corpus consisting of 15 different translations in 10 languages from the Christian Bible to evaluate our alignment model and error recovery methods. For example, based on noisy target language phoneme sequences with 45.1% errors, we build a dictionary for an English Bible with a Spanish Bible translation with 4.5% OOV rate, where 64% of the extracted pronunciations contain no more than one wrong phoneme. Finally, we use the extracted pronunciations in an automatic speech recognition system for the target language and report promising word error rates – given that pronunciation dictionary and language model are learned completely unsupervised and no written form for the target language is required for our approach.

Translation of Untranslatable Words-Integration of Lexical Approximation and Phrase-Table Extension Techniques into Statistical Machine Translation

This paper proposes a method for handling out-of-vocabulary (OOV) words that cannot be translated using conventional phrase-based statistical machine translation (SMT) systems. For a given OOV word, lexical approximation techniques are utilized to identify spelling and inflectional word variants that occur in the training data. All OOV words in the source sentence are then replaced with appropriate word variants found in the training corpus, thus reducing the number of OOV words in the input. Moreover, in order to increase the coverage of such word translations, the SMT translation model is extended by adding new phrase translations for all source language words that do not have a single-word entry in the original phrase-table but only appear in the context of larger phrases. The effectiveness of the proposed methods is investigated for the translation of Hindi to English, Chinese, and Japanese.

Frequency-based identification of correct translation equivalents (FITE) obtained through transformation rules

We devised a novel statistical technique for the identification of the translation equivalents of source words obtained by transformation rule based translation (TRT). The effectiveness of the technique called frequency-based identification of translation equivalents ( FITE ) was tested using biological and medical cross-lingual spelling variants and out-of-vocabulary (OOV) words in Spanish-English and Finnish-English TRT. The results showed that, depending on the source language and frequency corpus, FITE-TRT (the identification of translation equivalents from TRT's translation set by means of the FITE technique) may achieve high translation recall. In the case of the Web as the frequency corpus, translation recall was 89.2%--91.0% for Spanish-English FITE-TRT. For both language pairs FITE-TRT achieved high translation precision: 95.0%--98.8%. The technique also reliably identified native source language words: source words that cannot be correctly translated by TRT. Dictionary-based CLIR augmented with FITE-TRT performed substantially better than basic dictionary-based CLIR where OOV keys were kept intact. FITE-TRT with Web document frequencies was the best technique among several fuzzy translation/matching approaches tested in cross-language retrieval experiments. We also discuss the application of FITE-TRT in the automatic construction of multilingual dictionaries.

Automatic extraction of bilingual word pairs from parallel corpora with various languages using learning for adjacent information

AbstractThis paper presents a learning method using adjacent information as the method to extract bilingual word pairs efficiently from parallel corpora with various languages for which language resources are insufficient. In our method, information about correspondence between source language words and target language words is acquired automatically using the word strings that adjoin bilingual word pairs. That acquired information is used to solve the ambiguity problem of correspondence between source language words and target language words in various bilingual sentence pairs. First, the system using our method automatically acquires templates as information that indicates correspondence between source language words and target language words. The templates are based on word strings that adjoin the bilingual word pairs. Moreover, the system using our method efficiently extracts bilingual word pairs from bilingual sentence pairs using the acquired templates. Evaluation experiments showed that the system using our method extracted bilingual word pairs from parallel corpora with five kinds of languages. Results show that the total extraction rate was 60.1%. The total extraction rate was better by 8.0 percentage points compared to that obtained using a system based only on the Dice coefficient without our method. Those results confirm the effectiveness of our method. © 2006 Wiley Periodicals, Inc. Syst Comp Jpn, 37(13): 40–53, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.20534

`단어-의미 의미-단어` 관계에 기반한 번역어 선택

기계번역에서 올바른 번역 문장을 구성하기 위해서는 원시 문장의 의미를 올바르게 표현하면서 자연스러운 목적 문장을 구성하는 번역어를 선택해야 한다. 본 논문에서는 '단어-의미 의미-단어' 관계, 즉 원시언어의 한 단어는 하나 이상의 의미를 가지고 각 의미는 각기 다른 목적언어 단어로 표현된다는 점에 기반하여, 원시 단어의 의미 분별과 목적 단어 선택을 결합하여 번역어를 선택하는 방식을 제안한다. 기존의 번역방식은 원시 단어에 대한 목적단어를 직접 선택하는 '단어-단어' 관계에 기반하고 있기 때문에, 원시언어를 목적 언어로 직접 대응시키기 위한 지식을 필요로 하여 지식 획득에 어려움이 있었다. 본 논문의 방식에서는 원시 단어의 의미 분별과 목적 언어의 단어 선택의 결합을 통해 번역어를 선택함으로써, 손쉽게 획득할 수 있는 원시 언어와 목적 언어 각각의 지식원에서 번역어 선택을 위한 지식을 자동으로 추출할 수 있다. 또한 원시 언어의 의미와 목적 언어의 쓰임새를 모두 반영하여 충실도와 이해도를 모두 만족시키는 보다 정확한 번역어를 선택할 수 있다. To obtain a correctly translated sentence in a machine translation system, we must select target words that not only reflect an appropriate meaning in a source sentence but also make a fluent sentence in a target language. This paper points out that a source language word has various senses and each sense can be mapped into multiple target words, and proposes a new translation disambiguation method based on this 'word-to-sense and sense-to-word' relationship. In my method target words are chosen through disambiguation of a source word sense and selection of a target word. Most of translation disambiguation methods are based on a 'word-to-word' relationship that means they translate a source word directly into a target wort so they require complicate knowledge sources that directly link a source words to target words, which are hard to obtain like bilingual aligned corpora. By combining two sub-problems for each language, knowledge for translation disambiguation can be automatically extracted from knowledge sources for each language that are easy to obtain. In addition, disambiguation results satisfy both fidelity and intelligibility because selected target words have correct meaning and generate naturally composed target sentences.

Translation Selection by Combining Multiple Measures for Sense Disambiguation and Word Selection

Translation selection is a process to select, from a set of target language words corresponding to a source language word, the most appropriate one that conveys the correct sense of a source word and makes the target language sentence more natural. In this paper, we propose a hybrid method for translation selection that exploits a bilingual dictionary and a target language corpus. Based on the 'word-to-sense and sense-to-word' relationship between a source word and its translations, our method selects translation through two levels: sense disambiguation of a source word and selection of a target word. For translation selection, we introduce three measures: sense preference and sense probability for sense disambiguation, and word probability for word selection. The first one is based on knowledge from a bilingual dictionary, and the others are calculated using statistics from a target language corpus. We evaluated our method and results showed that our method selects more appropriate target words with knowledge extracted from easily obtainable resources.

Defining Linear Context to Resolve Lexical Ambiguity

This paper sets out a mechanical procedure for eliminating ambiguities in machine translation in cases where a word in the source language has a number of equivalents in the target language. The procedure depends on the inspection of linear word context and the establishing of “diagnostic environments” for the ambiguous words. In the experiment described, the context was restricted to one word before and one following the ambiguous word. It demonstrated that considerable reduction in the number of ambiguities can be achieved by the procedure and that the word following the ambiguous word is more effective than the word preceding in resolving the ambiguity.