Skip to main content

BUSQUEDA DE INFORMACION EN CONTENIDOS AUDIOVISUALES PLURILINGUES

TEC2009-14094-C04-01

•
Funding agency name Ministerio de Ciencia e Innovación
Funding agency acronym MICINN
Program Programa Nacional de Proyectos de Investigación Fundamental
Subprogram Subprograma de Proyectos de Investigación Fundamental no Orientada
Call Proyectos I+D
Call year 2009
State Plan 2008-2011 Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica
Management unit Subdirección General de Proyectos de Investigación
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 7
Encontrada(s) 1 página(s)

Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Automatic and human evaluation study of a rule-based and a statistical Catalan-Spanish machine translation systems

  • Costa-jussà, Marta R.
  • Farrús, Mireia
  • Mariño Acebal, José B.
  • Fonollosa, José A Rodriguez
Comunicació presentada a: LREC 2010, Seventh International Conference on Language Resources and Evaluation, celebrat a Valletta, Malta, del 17 al 23 de maig de 2010., Machine translation systems can be classified into rule-based and corpus-based approaches, in terms of their core technology. Since both paradigms have largely been used during the last years, one of the aims in the research community is to know how these systems differ in terms of translation quality. To this end, this paper reports a study and comparison of a rule-based and a corpus-based (particularly, statistical) Catalan-Spanish machine translation systems, both of them freely available in the web. The translation quality analysis is performed under two different domains: journalistic and medical. The systems are evaluated by using standard automatic measures, as well as by native human evaluators. Automatic results show that the statistical system performs better than the rule-based system. Human judgements show that in the Spanishto-Catalan direction the statistical system also performs better than the rule-based system, while in the Catalan-to-Spanish direction is the other way round. Although the statistical system obtains the best automatic scores, its errors tend to be more penalized by human judgements than the errors of the rule-based system. This can be explained because statistical errors are usually unexpected and they do not follow any pattern., This work has been partially funded by the Spanish Department of Education and Science through the Juan de la Cierva fellowship program and the BUCEADOR project (TEC2009-14094-C04-01). The authors also wants to thank the Barcelona Media Innovation Centre for its support and permission to publish this research and the UniversalDoctor Project for providing the medical test material.
Repositorio Digital de la UPF
Publication Artículo científico (JournalArticle)

Overcoming statistical machine translation limitations: error analysis and proposed solutions for the Catalan–Spanish language pair

  • Farrús, Mireia
  • Costa-jussà, Marta R.
  • Mariño Acebal, José B.
  • Poch, Marc
  • Hernández, Adolfo
  • Henríquez, Carlos
  • Fonollosa, José A Rodriguez
This work aims to improve an N-gram-based statistical machine translation system between the Catalan and Spanish languages, trained with an aligned Spanish-Catalan parallel corpus consisting of 1.7 million sentences taken from El Periódico newspaper. Starting from a linguistic error analysis above this baseline system, orthographic, morphological, lexical, semantic and syntactic problems are approached using a set of techniques. The proposed solutions include the development and application of additional statistical techniques, text pre- and post-processing tasks, and rules based on the use of grammatical categories, as well as lexical categorization. The performance of the improved system is clearly increased, as is shown in both human and automatic evaluations of the system, with a gain of about 1.1 points BLEU observed in the Spanish-to-Catalan direction of translation, and a gain of about 0.5 points in the reverse direction. The final system is freely available online as a linguistic resource., This work has been partially funded by the Spanish Department of Science and Innovation through the Juan de la Cierva fellowship program and the Spanish Government under the BUCEADOR project (TEC2009-14094-C04-01).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Automatic evaluation of continuous assessment tests

  • Farrús, Mireia
  • Costa-jussà, Marta R.
  • Cobo, Germán
  • García, David
  • Villarejo Muñoz, Luis
  • Banchs, Rafael E
Comunicació presentada a: 6th ICTATLL, celebrat a Kyoto, Japó, del 21-23 de setembre de 2010, The UOC is an online university where students follow an online continuous assessment system in order to successfully complete their studies. Due to the growing number of students, the task of marking their assessment tests has become tedious and time-consuming. In order to speed up and standardise the task, the UOC started to develop a pre-evaluation system to provide a preliminary, automatic evaluation of the tests by using the latent semantic analysis technique. Given a set of ideal reference answers for the evaluation tests, the students’ answers are evaluated with respect to the reference, providing an estimated score. This paper shows some preliminar results, from which we can conclude that the correlation between automatic and human evaluations is higher and statistically significant when both the math formulas codification and the language used by the students are the same than in the reference answers., This work has been partially funded by the Universitat Oberta de Catalunya under the Teaching Innovation Project number IN-PID1043 and by the Spanish Department of Education and Science through the Juan de la Cierva fellowship program and the BUCEADOR project number TEC2009-14094-C04-01. The authors also want to thank the Barcelona Media Innovation Centre for its support and permission to publish this research.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Automatic and human evaluation study of a rule-based and a statistical Catalan-Spanish machine translation systems

  • Costa-jussà, Marta R.
  • Farrús, Mireia
  • Mariño Acebal, José B.
  • Fonollosa, José A Rodriguez
Comunicació presentada a: LREC 2010, Seventh International Conference on Language Resources and Evaluation, celebrat a Valletta, Malta, del 17 al 23 de maig de 2010., Machine translation systems can be classified into rule-based and corpus-based approaches, in terms of their core technology. Since both paradigms have largely been used during the last years, one of the aims in the research community is to know how these systems differ in terms of translation quality. To this end, this paper reports a study and comparison of a rule-based and a corpus-based (particularly, statistical) Catalan-Spanish machine translation systems, both of them freely available in the web. The translation quality analysis is performed under two different domains: journalistic and medical. The systems are evaluated by using standard automatic measures, as well as by native human evaluators. Automatic results show that the statistical system performs better than the rule-based system. Human judgements show that in the Spanishto-Catalan direction the statistical system also performs better than the rule-based system, while in the Catalan-to-Spanish direction is the other way round. Although the statistical system obtains the best automatic scores, its errors tend to be more penalized by human judgements than the errors of the rule-based system. This can be explained because statistical errors are usually unexpected and they do not follow any pattern., This work has been partially funded by the Spanish Department of Education and Science through the Juan de la Cierva fellowship program and the BUCEADOR project (TEC2009-14094-C04-01). The authors also wants to thank the Barcelona Media Innovation Centre for its support and permission to publish this research and the UniversalDoctor Project for providing the medical test material.
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Overcoming statistical machine translation limitations: error analysis and proposed solutions for the Catalan–Spanish language pair

  • Farrús, Mireia
  • Costa-jussà, Marta R.
  • Mariño Acebal, José B.
  • Poch, Marc
  • Hernández, Adolfo
  • Henríquez, Carlos
  • Fonollosa, José A Rodriguez
This work aims to improve an N-gram-based statistical machine translation system between the Catalan and Spanish languages, trained with an aligned Spanish-Catalan parallel corpus consisting of 1.7 million sentences taken from El Periódico newspaper. Starting from a linguistic error analysis above this baseline system, orthographic, morphological, lexical, semantic and syntactic problems are approached using a set of techniques. The proposed solutions include the development and application of additional statistical techniques, text pre- and post-processing tasks, and rules based on the use of grammatical categories, as well as lexical categorization. The performance of the improved system is clearly increased, as is shown in both human and automatic evaluations of the system, with a gain of about 1.1 points BLEU observed in the Spanish-to-Catalan direction of translation, and a gain of about 0.5 points in the reverse direction. The final system is freely available online as a linguistic resource., This work has been partially funded by the Spanish Department of Science and Innovation through the Juan de la Cierva fellowship program and the Spanish Government under the BUCEADOR project (TEC2009-14094-C04-01).
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Automatic evaluation of continuous assessment tests

  • Farrús, Mireia
  • Costa-jussà, Marta R.
  • Cobo, Germán
  • García, David
  • Villarejo Muñoz, Luis
  • Banchs, Rafael E
Comunicació presentada a: 6th ICTATLL, celebrat a Kyoto, Japó, del 21-23 de setembre de 2010, The UOC is an online university where students follow an online continuous assessment system in order to successfully complete their studies. Due to the growing number of students, the task of marking their assessment tests has become tedious and time-consuming. In order to speed up and standardise the task, the UOC started to develop a pre-evaluation system to provide a preliminary, automatic evaluation of the tests by using the latent semantic analysis technique. Given a set of ideal reference answers for the evaluation tests, the students’ answers are evaluated with respect to the reference, providing an estimated score. This paper shows some preliminar results, from which we can conclude that the correlation between automatic and human evaluations is higher and statistically significant when both the math formulas codification and the language used by the students are the same than in the reference answers., This work has been partially funded by the Universitat Oberta de Catalunya under the Teaching Innovation Project number IN-PID1043 and by the Spanish Department of Education and Science through the Juan de la Cierva fellowship program and the BUCEADOR project number TEC2009-14094-C04-01. The authors also want to thank the Barcelona Media Innovation Centre for its support and permission to publish this research.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Evaluación de estrategias para la traducción automática estadística de chino a castellano con el inglés como lengua pivote, Evaluating indirect strategies for Chinese-Spanish statistical machine translation with English as pivot language

  • Ruiz Costa-Jussà, Marta
  • Henríquez Quintana, Carlos
  • Banchs Martínez, Rafael Enrique
El chino y el castellano son los idiomas más hablados en el mundo como lenguas maternas. Sin embargo, no existe mucha actividad de investigación en traducción automática entre este par de lenguas. Este artículo se enfoca en la investigación del estado actual de la cuestión de la traducción automática estadística entre chino-castellano, ya que hoy en día constituye una de las aproximaciones más usadas dentro del área de la traducción automática. Con este propósito en mente, describimos los corpus paralelos disponibles como el BTEC (Basic Traveller Expressions Corpora), la Biblia y las Naciones Unidas (UN). Concretamente, experimentamos con diferentes estrategias de traducción automática estadística directa e indirectas (denominadas pivotes). Entre las estrategias pivotes exploramos dos metodologías: la traducción de chino a pivote y de pivote a castellano; y el sistema entrenado con un pseudo-corpus chino-castellano, en el que el castellano se ha traducido previamente del pivote. Usamos el inglés como lengua pivote. Los resultados experimentales sugieren que el inglés podría constituir una lengua óptima para la intermediación de la traducción entre chino y castellano. Así pues, uno de los principales objetivos de este trabajo es motivar a la comunidad científica para investigar en este par de lenguas de alto impacto demográfico.