Towards incorporating language morphology into statistical machine translation systems

Karageorgakis, P.; Potamianos, Alexandros; Klasinas, Ioannis

doi:10.1109/asru.2005.1566533

Cited by 4 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many SMT systems today do not rely on linguistic knowledge and instead use a direct translation approach at the lexical level. [7] One of the primary causes of lexical mistranslations in SMT is the over-reliance on statistical probabilities to determine the most likely translation of a word or phrase. This approach can be effective for translating simple, straightforward sentences, but it often fails to capture the nuances and complexities of more complex prose text.…”

Section: Lexical Mistranslationmentioning

confidence: 99%

An Analysis of the Factors of Mistranslation in Statistical Machine Translation—Taking Prose Text Translation as Example

2023

AJHSS

View full text Add to dashboard Cite

This article analyzes the factors of mistranslation in Statistical Machine Translation (STM) bytaking prose translation as an example. First, it reviews the basics of STM and prose translation, and then discusses the common mistranslation types in STM, including lexical, syntactic, semantic, and pragmatic mistakes. Next, it identifies possible factors contributing to these mistranslations, including the quality of training data, the selected translation model, and the limitations of machine learning algorithms. Finally, it proposes some possible solutions to reduce mistranslation in STM, such as improving the quality of training data, selecting more appropriate translation models, and exploring new mistranslation detection techniques. The analysis presented in this article provides valuable insights into the challenges and opportunities in STM, and helps improve the accuracy and quality of machine translation.

show abstract

Section: Lexical Mistranslationmentioning

confidence: 99%

An Analysis of the Factors of Mistranslation in Statistical Machine Translation—Taking Prose Text Translation as Example

2023

AJHSS

View full text Add to dashboard Cite

show abstract

“…The challenges raised when translating from or into richer morphology languages are well known and are being continuously studied in the context of PBSMT. Morphology is the study of the structure of a set of given language morphemes, such as stems or affixes [Karageorgakis et al 2005]. Although the most important morpheme is the stem, in this paper we will deal with morphemes other than the stem.…”

Section: Morphologymentioning

confidence: 99%

Statistical machine translation enhancements through linguistic levels

Costa-jussà

Farrús

2014

ACM Comput. Surv.

View full text Add to dashboard Cite

Machine translation can be considered a highly interdisciplinary and multidisciplinary field because it is approached from the point of view of human translators, engineers, computer scientists, mathematicians and linguists. One of the most popular approaches is the statistical machine translation (SMT) approach which tries to cover translation in a holistic manner by learning from parallel corpus aligned at the sentence level. However, with this basic approach, there are some issues at each written linguistic level (i.e. orthographic, morphological, lexical, syntactic and semantic) that remain unsolved. Research in SMT has continuously been focused on solving the different linguistic levels challenges. This paper represents a survey of how the SMT has been enhanced to perform translation correctly at all linguistic levels.

show abstract

Segmentation Strategies to Face Morphology Challenges in Brazilian-Portuguese/English Statistical Machine Translation and Its Integration in Cross-Language Information Retrieval

Costa-jussà

2015

CyS

View full text Add to dashboard Cite

Abstract.The use of morphology is particularly interesting in the context of statistical machine translation in order to reduce data sparseness and compensate any lack of training corpus. In this work, we propose several approaches to introduce morphology knowledge into a standard phrase-based machine translation system. We provide word segmentation using two different tools (COGROO and MORFESSOR) which allow to reduce the vocabulary and data sparseness. Then, we add to these segmentations the morphological information of a POS language model. We combine all these approaches using a Minimum Bayes Risk strategy. Experiments show significant improvements from the enhanced system over the baseline system on Brazilian Portuguese/English language pair. Finally, we report a case study about the impact of enhancing the statistical machine translation system with morphology in a cross-language application system such as ONAIR which allows users to look for information in video fragments through queries in natural language. Keywords.Morphology, Factored-based Machine Translation, Cross-language Information Retrieval Estrategias de segmentación para afrontar los retos morfológicos en un sistema de traducción automática estadística y su integración en un sistema de búsqueda de información crosslingüe.Resumen. La segmentación basada en morfología es particularmente interesante en el contexto de la traducción automática estadística para reducir la dispersión de los datos y compensar la falta de los mismos. En este trabajo, proponemos diferentes aproximaciones para incorporar conocimiento morfológico en un sistema estándard de traducción automática estadística basado en segmentos. Propocionamos una segmentación en palabras usando dos herramientas diferentes (COGROO y MORFESSOR) que permiten precisamente reducir la dispersión de los datos. Después, añadimos información morfológica al modelo de lenguaje. Combinamos estas aproximaciones usando una estrategia para minimizar el riesgo de Bayes. Los experimentos demuestran mejoras significativas del sistema mejorado respecto al sistema de referencia en el par Brasileño/Inglés. Finalmente, reportamos un caso de estudio sobre el impacto de mejorar el sistema de traducción estadística con morfología en una aplicación como ONAIR que permite a los usuarios buscar información en fragmentos audiovisuales usando consultas en lenguaje natural.

show abstract

Towards incorporating language morphology into statistical machine translation systems

Cited by 4 publications

References 15 publications

An Analysis of the Factors of Mistranslation in Statistical Machine Translation—Taking Prose Text Translation as Example

An Analysis of the Factors of Mistranslation in Statistical Machine Translation—Taking Prose Text Translation as Example

Statistical machine translation enhancements through linguistic levels

Segmentation Strategies to Face Morphology Challenges in Brazilian-Portuguese/English Statistical Machine Translation and Its Integration in Cross-Language Information Retrieval

Contact Info

Product

Resources

About