Evaluation of machine translation output is an important task. Various human evaluation techniques as well as automatic metrics have been proposed and investigated in the last decade. However, very few evaluation methods take the linguistic aspect into account. In this article, we use an objective evaluation method for machine translation output that classifies all translation errors into one of the five following linguistic levels: orthographic, morphological, lexical, semantic, and syntactic. Linguistic guidelines for the target language are required, and human evaluators use them in to classify the output errors. The experiments are performed on English-to-Catalan and Spanish-to-Catalan translation outputs generated by four different systems: 2 rule-based and 2 statistical. All translations are evaluated using the 3 following methods: a standard human perceptual evaluation method, several widely used automatic metrics, and the human linguistic evaluation. Pearson and Spearman correlation coefficients between the linguistic, perceptual, and automatic results are then calculated, showing that the semantic level correlates significantly with both perceptual evaluation and automatic metrics.

@article{Farreus2012, author = {Farreús, Mireia and Costa jussà, Marta R and Morse, Maja Popović}, title = {Study and correlation analysis of linguistic, perceptual, and automatic machine translation evaluations}, journal = {Journal of the American Society for Information Science and Technology}, year = {2012}, volume = {63}, pages = {174--184}, url = {http://dx.doi.org/10.1002/asi.21674}, doi = {10.1002/asi.21674}, issn = {1532-2890}, keywords = {Análisis, Evaluac, Traducción, automática, lingüstico}, abstract = {Evaluation of machine translation output is an important task. Various human evaluation techniques as well as automatic metrics have been proposed and investigated in the last decade. However, very few evaluation methods take the linguistic aspect into account. In this article, we use an objective evaluation method for machine translation output that classifies all translation errors into one of the five following linguistic levels: orthographic, morphological, lexical, semantic, and syntactic. Linguistic guidelines for the target language are required, and human evaluators use them in to classify the output errors. The experiments are performed on English-to-Catalan and Spanish-to-Catalan translation outputs generated by four different systems: 2 rule-based and 2 statistical. All translations are evaluated using the 3 following methods: a standard human perceptual evaluation method, several widely used automatic metrics, and the human linguistic evaluation. Pearson and Spearman correlation coefficients between the linguistic, perceptual, and automatic results are then calculated, showing that the semantic level correlates significantly with both perceptual evaluation and automatic metrics.} }

%0 = article %A = Farreús, Mireia and Costa jussà, Marta R and Morse, Maja Popović %D = 2012 %T = Study and correlation analysis of linguistic, perceptual, and automatic machine translation evaluations %U = http://dx.doi.org/10.1002/asi.21674

The Government of Canada’s Translation Bureau (‘the Bureau’) engaged PricewaterhouseCoopers LLP (‘PwC’) between January 16 and May 15, 2012 to conduct a benchmarking and comparative analysis study. The objectives of the comparative analysis study were to provide: • Information and analysis on the capacity of the Canadian industry to meet national and government demand; and • Benchmarks for good practices in linguistic services with other organizations of comparable size at the national and international level.

@book{PWC2012, author = {PWC}, title = {Translation Bureau Benchmarking and Comparative Analysis : Final Report May 15, 2012}, publisher = {PWC}, address = {Otawa}, year = {2012}, url = {http://www.btb.gc.ca/publications/documents/rapport-report-benchmarking-eng.pdf}, keywords = {Calidad, Evaluac, Servicios, Traducción, de, traducción}, abstract = {The Government of Canada’s Translation Bureau (‘the Bureau’) engaged PricewaterhouseCoopers LLP (‘PwC’) between January 16 and May 15, 2012 to conduct a benchmarking and comparative analysis study. The objectives of the comparative analysis study were to provide: • Information and analysis on the capacity of the Canadian industry to meet national and government demand; and • Benchmarks for good practices in linguistic services with other organizations of comparable size at the national and international level.} }

%0 = book %A = PWC %C = Otawa %D = 2012 %I = PWC %T = Translation Bureau Benchmarking and Comparative Analysis : Final Report May 15, 2012 %U = http://www.btb.gc.ca/publications/documents/rapport-report-benchmarking-eng.pdf

En este texto se presentan algunos aspectos y hallazgos de la investigación: "Análisis crtico de los discursos evaluativos de los docentes de lengua castellana de los grados décimo (10º) y once (11º), de algunas instituciones educativas de Ibagué" (2007). Para llevar a cabo lo anterior, se expone, a manera de introducción, algunos aspectos generales del estudio discursivo; en forma seguida, se plantea la postura teórica empleada como eje investigativo y argumental de la interpretación y el análisis de las muestras. De igual manera, se presenta el análisis crtico de algunas muestras discursivas evaluativas orales y escritas, y en última instancia, las conclusiones y la bibliografa de referencia empleada.

@article{LondonoZapata2009, author = {Londoño Zapata, Oscar Iván}, title = {Una aproximación al análisis crtico de los discursos evalutativos (ACDE)}, journal = {Revista de Lingüstica y Lenguas Aplicadas}, year = {2009}, pages = {103--124}, url = {http://dialnet.unirioja.es/servlet/extart?codigo=2993225}, issn = {1886-2438}, keywords = {Análisis, Evaluac, Lingüstica, aplicada, del, discurso}, abstract = {En este texto se presentan algunos aspectos y hallazgos de la investigación: "Análisis crtico de los discursos evaluativos de los docentes de lengua castellana de los grados décimo (10º) y once (11º), de algunas instituciones educativas de Ibagué" (2007). Para llevar a cabo lo anterior, se expone, a manera de introducción, algunos aspectos generales del estudio discursivo; en forma seguida, se plantea la postura teórica empleada como eje investigativo y argumental de la interpretación y el análisis de las muestras. De igual manera, se presenta el análisis crtico de algunas muestras discursivas evaluativas orales y escritas, y en última instancia, las conclusiones y la bibliografa de referencia empleada.} }

%0 = article %A = Londoño Zapata, Oscar Iván %D = 2009 %T = Una aproximación al análisis crtico de los discursos evalutativos (ACDE) %U = http://dialnet.unirioja.es/servlet/extart?codigo=2993225