To see the other types of publications on this topic, follow the link: Post-editing.
Dissertations / Theses on the topic 'Post-editing'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Post-editing.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Blagodarna, Olena. "Enhancement of post-editing performance: introducing machine translation post-editing in translator training." Doctoral thesis, Universitat Autònoma de Barcelona, 2018. http://hdl.handle.net/10803/666847.
Full textAbstract:
Los objetivos clave de esta tesis fueron explorar el perfil de los traductores involucrados en posedición, establecer el alcance de las competencias y habilidades requeridas, y sugerir una propuesta de formación válida que mejore el rendimiento de los post-editores noveles de acuerdo con los requisitos del Espacio Europeo de Educación Superior.
La tesis integró dos estudios secuenciales: una investigación basada en una encuesta que proporcionó información real sobre los perfiles y prácticas de los post-editores, y una investigación empírico-experimental que puso a prueba el modelo de formación sugerido e involucró a un total de 46 estudiantes de traducción en el último año de su licenciatura. Para recoger pruebas concluyentes sobre la aplicabilidad de la propuesta en diferentes contextos lingüísticos, el estudio se centró en 22 participantes que eran estudiantes de la Universidad Aerospacial Nacional de Kharkiv (Ucrania) especializados en traducción inglés-ruso, y 24 participantes que eran estudiantes de la Universidad Autónoma de Barcelona (España) especializados en traducción inglés-español. El modelo de formación sugerido perseguía la adquisición de conocimientos conceptuales y operativos por parte de los participantes y se incorporó en un estudio experimental antes y después de la prueba. El impacto de dicho modelo se examinó mediante la evaluación de la calidad de los segmentos post-editados y las tasas de rendimiento demostradas por los participantes, así como las actitudes de los estudiantes hacia los temas relacionados con el traducción automática y posedición y la autoevaluación de su rendimiento antes y después de dicha formación. La tesis termina con una reflexión sobre los cambios que podrían introducirse en la formación si se utilizaran sistemas neurales de traducción automática para generar el corpus de formación.
La tesis contribuye a la definición del alcance de la experiencia profesional de los post-editores, ofrece un modelo de formación escalable y describe hasta qué punto dicho modelo puede mejorar la posedición de los estudiantes universitarios de traducción.The key objectives of this thesis were to explore the profile of translators involved in post-editing, to outline the scope of required competencies and skills and to suggest a valid training proposal that would enhance post-editing performance of novice post-editors in conformity with the European Higher Education Area requirements. The thesis integrated two sequential studies: a survey-based research that yielded authentic information concerning post-editors’ profiles and practices and an empirical-experimental research that put the suggested training model to the test and involved a total of 46 translation students in the final year of their Bachelor’s program. To collect conclusive evidence about the applicability of the proposal across different linguistic backgrounds, the study focused on 22 participants who were students at Kharkiv National Aerospace University (Ukraine) and specialized in English-Russian translation, and 24 participants who were students at Universitat Autònoma de Barcelona (Spain) and specialized in English-Spanish translation. The suggested training model pursued acquisition of conceptual and operational knowledge by the trainees and was incorporated in a pretest-posttest experimental study. The impact of such model was examined by the evaluation of the quality of post-edited segments and throughput rates demonstrated by the participants as well as the students’ attitudes to MTPE-related issues and self-evaluation of their post-editing performance before and after the training. The thesis ends with reflections upon the changes that might be brought to the proposal if neural machine translation systems were used to generate the training corpus. The dissertation contributes to the definition of the scope of post-editors’ professional expertise, offers a scalable training model and describes to what extent such model may enhance post-editing performance in undergraduate translation students.
2
Chatterjee, Rajen. "Automatic Post-Editing for Machine Translation." Doctoral thesis, Università degli studi di Trento, 2019. http://hdl.handle.net/11572/242495.
Full textAbstract:
Automatic Post-Editing (APE) aims to correct systematic errors in a machine translated text. This is primarily useful when the machine translation (MT) system is not accessible for improvement, leaving APE as a viable option to improve translation quality as a downstream task - which is the focus of this thesis. This field has received less attention compared to MT due to several reasons, which include: the limited availability of data to perform a sound research, contrasting views reported by different researchers about the effectiveness of APE, and limited attention from the industry to use APE in current production pipelines.
In this thesis, we perform a thorough investigation of APE as a down- stream task in order to: i) understand its potential to improve translation quality; ii) advance the core technology - starting from classical methods to recent deep-learning based solutions; iii) cope with limited and sparse data; iv) better leverage multiple input sources; v) mitigate the task-specific problem of over-correction; vi) enhance neural decoding to leverage external knowledge; and vii) establish an online learning framework to handle data diversity in real-time.
All the above contributions are discussed across several chapters, and most of them are evaluated in the APE shared task organized each year at the Conference on Machine Translation. Our efforts in improving the technology resulted in the best system at the 2017 APE shared task, and our work on online learning received a distinguished paper award at the Italian Conference on Computational Linguistics. Overall, outcomes and findings of our work have boost interest among researchers and attracted industries to examine this technology to solve real-word problems.
3
Chatterjee, Rajen. "Automatic Post-Editing for Machine Translation." Doctoral thesis, Università degli studi di Trento, 2019. http://hdl.handle.net/11572/242495.
Full textAbstract:
Automatic Post-Editing (APE) aims to correct systematic errors in a machine translated text. This is primarily useful when the machine translation (MT) system is not accessible for improvement, leaving APE as a viable option to improve translation quality as a downstream task - which is the focus of this thesis. This field has received less attention compared to MT due to several reasons, which include: the limited availability of data to perform a sound research, contrasting views reported by different researchers about the effectiveness of APE, and limited attention from the industry to use APE in current production pipelines.
In this thesis, we perform a thorough investigation of APE as a down- stream task in order to: i) understand its potential to improve translation quality; ii) advance the core technology - starting from classical methods to recent deep-learning based solutions; iii) cope with limited and sparse data; iv) better leverage multiple input sources; v) mitigate the task-specific problem of over-correction; vi) enhance neural decoding to leverage external knowledge; and vii) establish an online learning framework to handle data diversity in real-time.
All the above contributions are discussed across several chapters, and most of them are evaluated in the APE shared task organized each year at the Conference on Machine Translation. Our efforts in improving the technology resulted in the best system at the 2017 APE shared task, and our work on online learning received a distinguished paper award at the Italian Conference on Computational Linguistics. Overall, outcomes and findings of our work have boost interest among researchers and attracted industries to examine this technology to solve real-word problems.
4
Housley, Jason K. "Ruqual: A System for Assessing Post-Editing." BYU ScholarsArchive, 2012. https://scholarsarchive.byu.edu/etd/3106.
Full textAbstract:
Post-editing machine translation has become more common in recent years due to the increase in materials requiring translation and the effectiveness of machine translation systems. This project presents a system for formalizing structured translation specifications that facilitates the assessment of the performance of a post-editor. This report provides details concerning two software applications: the Ruqual Specifications Writer, which aids in the authoring of post-editing project specifications, and the Ruqual Rubric Viewer which provides a graphical user interface for filling out a machine readable rubric file. The project as a whole relies on a definition of translation quality based on the specification approach. In order to test whether potential evaluators are able to reliably assess the quality of post-edited translations, a user study was conducted that utilized the Specifications Writer and Rubric Viewer. The specifications developed for the project were based on actual post-editing data provided by Ray Flournoy of Adobe. The study involved simulating the work of five post-editors, which consisted of developing texts and scenarios. 17 non-expert graders rated the work of the five fictional post-editors, and an Intraclass Correlation of the graders responses shows that they are reliable to a high degree. The groundwork laid by this project should help in the development of other applications that assist in the assessment of translation projects in terms of a specification approach to quality.
5
Sagemo, Oscar. "Estimating Post-Editing Effort with Translation Quality Features." Thesis, Uppsala universitet, Institutionen för lingvistik och filologi, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-299143.
Full text
6
Álvarez, Vidal Sergi. "Post-editing effort and linguistically motivated evaluation of machine translation." Doctoral thesis, Universitat Pompeu Fabra, 2021. http://hdl.handle.net/10803/672723.
Full textAbstract:
The recent improvements in neural MT (NMT) have driven a shift from
statistical MT (SMT) to NMT, which has propelled the use of post-editing
(PE) in translation workflows. However, many professional translators
state that if the quality of the MT output is not good enough, they delete
the remaining segments and translate everything from scratch. The problem
is that usual automatic measurements do not always indicate the quality
of theMT output, especially with high quality outputs, and there is still
no clear correlation between PE effort and productivity scores.
We combine quantitative and qualitative methods to study some of the
usual automatic metrics used to evaluate the quality of MT output, and
compare them to measures of post-editing effort. Then, we study in detail
different direct and indirect measures of effort in order to establish a
correlation among them. We complement this study with the analysis of
translators’ perceptions of the task.
Finally, we conduct a fine-grained analysis of MT errors based on postediting
corrections and suggest an error-based approach to evaluate raw
MT output which includes the use of challenge sets.Les recents millores en traducció automàtica neuronal (TAN) han provocat un canvi de la traducció automàtica estadística (TAS) a la TAN, que ha incrementat l’ús de la postedició en els fluxos de treball industrials. Tanmateix, molts traductors professionals afirmen que si la qualitat de la TA no és prou bona, eliminen tot el segment i el tradueixen de cap i de nou. El problema és que les mesures automàtiques habituals no sempre indiquen la qualitat de la TA, especialment quan aquesta és bona, i no hi ha una correlació directa entre l’esforç de postedició i les mesures de productivitat. Combinem mètodes quantitatius i qualitatius per estudiar algunes de les mesures automàtiques més habituals utilitzades per evaluar la qualitat de la TA, i les comparem amb les mesures de l’esforç de postedició. A continuació, estudiem en detall diferent mesures directes i indirectes d’esforç per establir-hi una correlació. Complementem aquest estudi amb l’anàlisi de les percepcions dels traductors que duen a terme aquesta tasca. Finalment, fem una anàlisis detallada dels errors de TA a partir de les correccions fetes en la postedició i proposem un enfocament basat en errors per evaluar la TA que inclou l’ús d’un conjunt de frases de prova.
7
Guerberof, Arenas Ana. "Productivity and quality in the post-editing of outputs from translation memories and machine translation." Doctoral thesis, Universitat Rovira i Virgili, 2012. http://hdl.handle.net/10803/90247.
Full textAbstract:
This study presents empirical research on no-match, machine-translated and translation-memory segments, analyzed in terms of translators’ productivity, final quality and prior professional experience. The findings suggest that translators have higher productivity and quality when using machine-translated output than when translating on their own, and that the productivity and quality gained with machine translation are not significantly different from the values obtained when processing fuzzy matches from a translation memory in the 85-94 percent range. The translators’ prior experience impacts on the quality they deliver but not on their productivity. These quantitative findings are triangulatedwith qualitative data from an online questionnaire and from one-to-one debriefings with the translators.
Este estudio presenta una investigación empírica sobre la traducción de segmentos nuevos y aquellos procesados con traducción automática y memorias de traducción analizados en relación a la productividad, calidad final y experiencia profesional de un grupo de traductores. Los resultados sugieren que los traductores obtienen una productividad y calidad más altas cuando procesan segmentos de traducción automática que cuando traducen sin ninguna ayuda y que dicha productividad y calidad no son significativamente diferentes a la que se obtiene cuando procesan coincidencias parciales de una memoria de traducción (del 85 al 94 por ciento). La experiencia profesional previa de los traductores influye en la calidad pero no así en la productividad obtenidas. Los resultados cuantitativos se triangulan, además, con datos cualitativos obtenidos a través de un cuestionario en línea y de entrevistas individuales realizadas a los trad
8
Ginovart, Cid Clara. "The need for practice in the acquisition of the post-editing skill-set: lessons learned from the industry." Doctoral thesis, Universitat Pompeu Fabra, 2021. http://hdl.handle.net/10803/672661.
Full textAbstract:
While (translation) technology and the language industry continue to
evolve, varieties of professional profiles and nuances to the relevant skillsets
emerge. The field of machine translation post-editing, and, in particular,
research on the learnability of its skills, has become one important focus of
attention.
With an exploratory sequential mixed-methods design, this dissertation
seeks to attain two goals. First, in the survey-research, the identification
of errors, the decision-making, and the respect of guidelines are confirmed
as core post-editing skills from an efficient editing standpoint. Second, in
the experimental study based on a pretest-posttest method, we evaluate the
treatment, that is, a proposal for a French-Spanish split-training course on
the three mentioned skills. The data analyses indicate positive results from
the quality perspective and suggest that longer practice is required to reach
proceduralisation.Mientras las tecnologías (de la traducción) y la industria lingüística siguen evolucionando, emergen variedades de perfiles profesionales cuyas habilidades se ven matizadas. El campo de la posedición de traducción automática y, en particular, la investigación en lo que respecta a la adquisición de las habilidades adecuadas, se han convertido en focos de atención primordiales. Con una metodología secuencial exploratoria mixta, en esta tesis se plantean dos objetivos. El primero, confirmar las principales habilidades de posedición desde una perspectiva de edición eficiente gracias a tres encuestas. Se concluye que son la identificación de errores, la toma de decisiones, y la aplicación de directrices. El segundo, evaluar una propuesta de formación fragmentada (split-training) en el par de lenguas francés-castellano sobre las tres habilidades mencionadas mediante un estudio experimental con método pretest-postest. Los resultados son positivos en cuanto a calidad traductológica, y sugieren que con más práctica y mayor experiencia se puede llegar a un grado de proceduralización adecuado en el ámbito profesional.
Mentre les tecnologies (de la traducció) i la indústria lingüística segueixen evolucionant, emergeixen varietats de perfils professionals i les habilitats que se’ls vinculen es van matisant. El camp de la post-edició de traducció automàtica i, en particular, la recerca sobre l’adquisició de les habilitats pertinents, s’han convertit en focus d’atenció cabdals. Amb una metodologia seqüencial exploratòria mixta, aquesta tesi té dos objectius. En primer lloc, la recerca mitjançant enquestes permet confirmar les principals habilitats de post-edició des d’una perspectiva d’edició efficient: la identificació d’errors, la presa de decisions, i l’aplicació de directrius. En segon lloc, l’estudi experimental basat en el mètode pretestpostest avalua el tractament, és a dir, la proposta d’una formació fragmentada (split-training) aplicada al parell d’idiomes francès-castellà sobre les tres habilitats esmentades. L’anàlisi conclou resultats positius quant a qualitat i suggereix que, per assolir un grau de proceduralització més adient, cal incrementar la pràctica.
9
Ortiz, Boix Carla. "Implementing machine translation and post-editing to the translation of wildlife documentaries through voice-over and off-screen dubbing." Doctoral thesis, Universitat Autònoma de Barcelona, 2016. http://hdl.handle.net/10803/400020.
Full textAbstract:
Aquesta tesi doctoral explora la possibilitat d’incloure la traducció automàtica i la traducció automàtica seguida de postedició en el procés de traducció de documentals científics de fauna i flora mitjançant veus superposades i doblatge en off. Per a aconseguir-ho, es van portar a terme tres estudis:
El primer estudi investiga els reptes que hauríem de superar per a poder introduir la traducció automàtica en el procés de traducció de documentals. Ens basem en estudis anteriors sobre traducció mitjançant veus superposades i doblatge en off i sobre traducció de documentals (per exemple, Franco et al., 2010) per a determinar els esmentats reptes, així com en l’anàlisi d’un corpus de guions de documentals en anglès, un corpus de documentals en castellà i un corpus de segments de guions de documentals i 9 traduccions, 8 de les quals van ser fetes per diversos motors de traducció automàtica en línia i gratuïts, i l’altra va ser traduïda per un professional. A més, l’estudi introdueix possibles solucions als reptes, que fem servir a l’experiment que presentem en el segon estudi.
El segon estudi se centra en un experiment per a determinar l’esforç necessari per a posteditar documentals científics de fauna i flora en comparació amb l’esforç requerit per a traduir-los. Dotze estudiants de màster van participar en l’experiment, que es va basar en la noció d’esforç de Krings (2001), així com en altres estudis sobre esforç de postedició (per exemple, O’Brien, 2006). Les conclusions de l’experiment indiquen que post-editar requereix menys esforç que traduir.
El tercer estudi compara la qualitat de documentals post-editats i traduïts. Per a fer-ho, es van portar a terme dos experiments. En el primer, sis experts van avaluar la qualitat dels documentals que van fer els estudiants de màster durant l’experiment del segon estudi. Així, doncs, els experts van posar nota als documents, els van corregir mitjançant una classificació d’errors basada en MQM que inclou especificacions per a la traducció documentals, i van respondre qüestionaris d’acord amb la seva opinió sobre les traduccions corregides. Les conclusions de l’experiment indiquen que la qualitat de les post-edicions i de les traduccions són significativament semblants.
En el segon experiment, 56 possibles usuaris van avaluar la qualitat dels documentals post-editats i traduïts mitjançant un qüestionari de recepció. Igual com va indicar l’avaluació dels experts, els resultats mostra que no hi ha diferencies significatives entre la qualitat dels documentals traduïts i els posteditats.
Els estudis presentats en aquesta tesi doctoral, doncs, permeten aconseguir l’objectiu principal de la tesi: “investigar si es podria incloure, satisfactòriament, la traducció automàtica, en quant a esforç i qualitat, en el procés de traducció de documentals d’un subdomini específic (fauna i flora) mitjançant veus superposades i doblatge en off; i validar les dues hipòtesis principals de la tesi: “esperem que la inclusió de traducció automàtica en el procés de traducció de documentals de naturalesa mitjançant veus superposades i doblatge en off optimitzi el procés en quant a esforç” i “esperem que la inclusió de traducció automàtica en el procés de traducció de documentals científics de fauna i flora per veus superposades i doblatge en off no afecti significativament la qualitat del producte traduït”.Esta tesis doctoral explora la posibilidad de incluir traducción automática y traducción automática seguida de post-edición en el proceso de traducción de documentales científicos de fauna y flora mediante voces superpuestas y doblaje en off. Para conseguirlo, llevamos a cabo tres estudios: El primer estudio investiga los desafíos que necesitamos superar para poder incluir la traducción automática en el proceso de traducción de documentales. Nos basamos en estudios anteriores sobre traducción mediante voces superpuestas y doblaje en off y sobre traducción de documentales (por ejemplo, Franco et al., 2010) para determinar los desafíos que nos podemos encontrar. También nos basamos en el análisis de un corpus de guiones de documentales en inglés, un corpus de documentales en español y un corpus de segmentos de guiones de documentales y sus nueve traducciones, ocho de las cuales fueron producidas por motores de traducción automática gratuitos y en línea, y otra traducción hecha por un traductor humano. Además, el estudio introduce posibles soluciones para estos desafíos, que usamos en el experimento presentado en el segundo estudio. El segundo estudio se centra en un experimento que pretende determinar el esfuerzo requerido para post-editar documentales de naturaleza comparado con el esfuerzo necesario para traducirlos. Doce estudiantes de máster participaron en el experimento, que se basó en la noción de esfuerzo de Krings (2001), así como otros estudios relacionados con esfuerzo de post-edición (por ejemplo, O’Brien, 2006). Las conclusiones del experimento parecen indicar que post-editar requiere menos esfuerzo que traducir. El tercer estudio compara la calidad de documentales post-editados y traducidos. Para hacerlo, se llevaron a cabo dos experimentos. En el primero, seis expertos evaluaron la calidad de los documentales hechos por los estudiantes durante el experimento: los expertos pusieron nota a los documentos, los corrigieron usando una clasificación de errores basada en el MQM que incluye especificaciones para la traducción de documentales, y respondieron unos cuestionarios sobre su opinión de las traducciones. Las conclusiones del experimento indican que la calidad de las post-ediciones y las traducciones son significativamente similares. En el segundo experimento, 56 posibles usuarios evaluaron la calidad de documentales post-editados y de documentales traducidos mediante un cuestionario de recepción. Igual como indicó la evaluación de los expertos, los resultados muestran que no hay diferencias significativas entre la calidad de los documentales traducidos y los post-editados. Los estudios presentados en esta tesis doctoral, pues, nos permiten conseguir el objetivo principal de la tesis: “investigar si se podría incluir la traducción automática de manera satisfactoria, en cuanto a esfuerzo y calidad, en el proceso de traducción de documentales de un subdominio específico (fauna y flora) mediante voces superpuestas y doblaje en off”; y validar las dos hipótesis principales de la tesis: “la inclusión de traducción automática en el proceso de traducción de documentales científicos de fauna y flora mediante voces superpuestas y doblaje en off optimizará el proceso en cuanto a esfuerzo” y “la inclusión de traducción automática en el proceso de traducción de documentales científicos de fauna y flora mediante voces superpuestas y doblaje en off no afectará significativamente la calidad del producto traducido”.
This PhD explores the possibility of introducing MT and PE MT into the process of translating wildlife documentaries to be voiced-over and off-screen dubbed. In order to do so, 3 studies have been carried out: The first stud researches the challenges that would need to be overcome if MT is to be included in the process of translating wildlife documentaries through VO and OD. The challenges are based on previous studies on the translation via VO and OD, and the translation of documentaries (e.g. Franco et al., 2010), as well as an analysis of a corpus of wildlife documentary scripts in English, a corpus of wildlife documentary scripts in Spanish, and a corpus of segments of documentary scripts and 9 translations: 8 produced by free-online MT engines and 1 by a human translator. Furthermore, it presents possible solutions to such challenges, which are used in the experiment presented in the second study. The second study presents and experiment that intends to determine the effort required to post-edit wildlife documentary films, as compared to the effort needed to translate them from scratch. 12 MA students participated in the experiment, which was based in Krings (2001) notion of PE effort, and previous experiments on PE effort (e.g. O’Brien, 2006). The conclusions of the experiments seem to indicate that PE requires less effort than translating them from scratch. The third study compares the quality of post-edited and translated wildlife documentaries. In order to do so, two experiments were carried out. The first one presents the QA, made by 6 experts, of the documentary scripts translated by MA students during the second study. The experts assessed the quality by grading the documents, by correcting them using an MQM-based error classification that includes specifications of documentary translation, and finally, by answering questionnaires on their opinions. Conclusions indicate that the quality of post-edited and translated wildlife documentaries is significantly similar. In the second experiment of the third study, 56 end-users blindly assessed the quality of post-edited and translated wildlife documentaries by using user-reception questionnaires. As well as the QA by experts, the results show that there is no significant difference between the quality of post-edited and translated documentaries. All the studies presented in this dissertation help to accomplish its main objective: “research whether MT might be successfully included, effort and quality wise, into the process of translating documentaries of a certain subdomain (wildlife) through VO and OD”; and validate its main hypotheses, which stand that “the inclusion of MT into the process of translating wildlife documentaries through VO and OD will optimize the process in terms of effort” and that “the inclusion of MT into the process of translating wildlife documentaries through VO and OD will not have significant impact on the quality of the translated product”.
10
Nunes, Vieira Lucas. "Cognitive effort in post-editing of machine translation : evidence from eye movements, subjective ratings, and think-aloud protocols." Thesis, University of Newcastle upon Tyne, 2016. http://hdl.handle.net/10443/3130.
Full textAbstract:
This thesis investigates the expenditure of cognitive effort in post-editing of machine translation. A mixed-method approach involving the use of eye movements, subjective ratings and think-aloud protocols was adopted for the investigation. The project aims at revealing connections between cognitive effort and variables including linguistic characteristics of the source text and the machine-translation output, post-editors’ individual traits, different linguistic aspects of the activity attended to during the task, and the overall quality of the post-edited text, assessed by human translators in terms of fluency (linguistic quality) and adequacy (faithfulness to the source text). Two tasks were conducted to pursue these aims: one involving eye tracking and a self-report scale of cognitive effort, and another carried out by a different, but comparable, sample of participants, under a think-aloud condition. Results indicate that variables such as an automatic machine-translation quality score and source-text type-token ratio are good predictors of cognitive effort in post-editing. The relationship between cognitive effort and post-editors’ traits was found to be a complex one, with significant links in this respect only appearing in the context of interactions between variables. A complex relationship was also found between editing behaviour and the quality of the post-edited text: the amount of changes implemented was found to have a generally positive association with post-edited fluency, though cognitive effort was found to be negatively correlated with both the fluency and adequacy of the post-edited text. Mental processes involving grammar and lexis were significantly related to the levels of cognitive effort expended by participants, being also the aspects most frequently attended to in the activity. From a methodological perspective, despite the criticisms received by the think-aloud method in previous research, empirical data obtained in this project indicate that think-aloud protocols correlate with eye movements and subjective ratings as measures of cognitive effort.
11
Ditchfield, Hannah. "Behind the screens of Facebook : an interactional study of pre-post editing and multicommunication in online social interaction." Thesis, University of Leicester, 2018. http://hdl.handle.net/2381/43051.
Full textAbstract:
This thesis explores the construction of posting activity within peer interactions on the social networking site, Facebook. At the heart of the thesis is an interest in how Facebook users negotiate online activities that are not possible in face-to-face, or other, communicative contexts. Two such activities are examined: pre-post editing and multicommunicating. Each of the activities explored in this thesis presents challenges to Facebook users. Users, for example, are presented with the unique, yet complex, ability to craft and re-craft their messages before sharing them with their audience as well as managing the interactional difficulties associated with engaging in multiple interactions at the same time. With such activities changing the dynamics of online interaction, this thesis explores how users utilise and manage these activities within their Facebook interactions as well as questioning the extent to which 'interactional order' is maintained. This thesis addresses these concerns by examining real time video recordings of four Facebook user's interactions that were generated through the use of screen-capture technology. Informed theoretically by the work of Goffman and methodologically by conversation analysis, this thesis goes 'beyond the screen' to examine the maintenance of interactional order within this pre-post activity. In doing so, the thesis makes a number of original contributions to knowledge relating to the study of online communication; presenting a unique perspective on how Facebook users maintain 'face' in pre-post interaction, exploring the use of "simplification techniques" within multicommunication activity and contributing to existing understandings of temporal organisation within online communicative environments. The thesis also makes a series of distinctive methodological interventions; challenging existing understandings of the public/private distinction in writing on online research ethics and working through methodological challenges of using screen capture software that have not yet been confronted by scholars.
12
Rossetti, Alessandra. "Perceived difficulty and time effectiveness in translation: a comparison of machine translation post-editing and translation memory use." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2014. http://amslaurea.unibo.it/7752/.
Full textAbstract:
Con il presente studio si è inteso analizzare l’impatto dell’utilizzo di una memoria di traduzione (TM) e del post-editing (PE) di un output grezzo sul livello di difficoltà percepita e sul tempo necessario per ottenere un testo finale di alta qualità. L’esperimento ha coinvolto sei studenti, di madrelingua italiana, del corso di Laurea Magistrale in Traduzione Specializzata dell’Università di Bologna (Vicepresidenza di Forlì). I partecipanti sono stati divisi in tre coppie, a ognuna delle quali è stato assegnato un estratto di comunicato stampa in inglese. Per ogni coppia, ad un partecipante è stato chiesto di tradurre il testo in italiano usando la TM all’interno di SDL Trados Studio 2011. All’altro partecipante è stato chiesto di fare il PE completo in italiano dell’output grezzo ottenuto da Google Translate. Nei casi in cui la TM o l’output non contenevano traduzioni (corrette), i partecipanti avrebbero potuto consultare Internet. Ricorrendo ai Think-aloud Protocols (TAPs), è stato chiesto loro di riflettere a voce alta durante lo svolgimento dei compiti. È stato quindi possibile individuare i problemi traduttivi incontrati e i casi in cui la TM e l’output grezzo hanno fornito soluzioni corrette; inoltre, è stato possibile osservare le strategie traduttive impiegate, per poi chiedere ai partecipanti di indicarne la difficoltà attraverso interviste a posteriori. È stato anche misurato il tempo impiegato da ogni partecipante. I dati sulla difficoltà percepita e quelli sul tempo impiegato sono stati messi in relazione con il numero di soluzioni corrette rispettivamente fornito da TM e output grezzo. È stato osservato che usare la TM ha comportato un maggior risparmio di tempo e che, al contrario del PE, ha portato a una riduzione della difficoltà percepita. Il presente studio si propone di aiutare i futuri traduttori professionisti a scegliere strumenti tecnologici che gli permettano di risparmiare tempo e risorse.
13
Scansani, Randy. "La traduzione automatica e i language service provider: il caso di CTI." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2016. http://amslaurea.unibo.it/9832/.
Full textAbstract:
La presente tesi nasce da un tirocinio avanzato svolto presso l’azienda CTI (Communication Trend Italia) di Milano. Gli obiettivi dello stage erano la verifica della possibilità di inserire gli strumenti automatici nel flusso di lavoro dell’azienda e l'individuazione delle tipologie testuali e delle combinazioni linguistiche a cui essi sono applicabili. Il presente elaborato si propone di partire da un’analisi teorica dei vari aspetti legati all’utilizzo della TA, per poi descriverne l’applicazione pratica nei procedimenti che hanno portato alla creazione dei sistemi custom.
Il capitolo 1 offre una panoramica teorica sul mondo della machine translation, che porta a delineare la modalità di utilizzo della TA ad oggi più diffusa: quella in cui la traduzione fornita dal sistema viene modificata tramite post-editing oppure il testo di partenza viene ritoccato attraverso il pre-editing per eliminare gli elementi più ostici.
Nel capitolo 2, partendo da una panoramica relativa ai principali software di traduzione automatica in uso, si arriva alla descrizione di Microsoft Translator Hub, lo strumento scelto per lo sviluppo dei sistemi custom di CTI.
Nel successivo passaggio, l’attenzione si concentra sull’ottenimento di sistemi customizzati. Un ampio approfondimento è dedicato ai metodi per reperire ed utilizzare le risorse. In seguito viene descritto il percorso che ha portato alla creazione e allo sviluppo dei due sistemi Bilanci IT_EN e Atto Costitutivo IT_EN in Microsoft Translator Hub.
Infine, nel quarto ed ultimo capitolo gli output che i due sistemi forniscono vengono rivisti per individuarne le caratteristiche e analizzati tramite alcuni tool di valutazione automatica. Grazie alle informazioni raccolte vengono poi formulate alcune previsioni sul futuro uso dei sistemi presso l’azienda CTI.
14
Mostofian, Nasrin. "A Study on Manual and Automatic Evaluation Procedures and Production of Automatic Post-editing Rules for Persian Machine Translation." Thesis, Uppsala universitet, Institutionen för lingvistik och filologi, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-325818.
Full textAbstract:
Evaluation of machine translation is an important step towards improving MT. One way to evaluate the output of MT is to focus on different types of errors occurring in the translation hypotheses, and to think of possible solutions to fix those errors. An error categorization is a rather beneficent tool that makes it easy to analyze the translation errors and can also be utilized to manually generate post-editing rules to be applied automatically to the product of machine translation. In this work, we define a categorization for the errors occurring in Swedish--Persian machine translation by analyzing the errors that occur in three data-sets from two websites: 1177.se, and Linköping municipality. We define three types of monolingual reference free evaluation (MRF), and use two automatic metrics BLEU and TER, to conduct a bilingual evaluation for Swedish-Persian translation. Later on, based on the experience of working with the errors that occur in the corpora, we manually generate automatic post-editing (APE) rules and apply them to the product of machine translation. Three different sets of results are obtained: (1) The results of analyzing MT errors show that the three most common types of errors that occur in the translation hypotheses are mistranslated words, wrong word order, and extra prepositions. These types of errors are placed in semantic and syntactic categories respectively. (2) The results of comparing the correlation between the automatic and manual evaluation show a low correlation between the two evaluations. (3) Lastly, applying the APE rules to the product of machine translation gives an increase in BLEU score on the largest data-set while remaining almost unchanged on the other two data-sets. The results for TER show a better score on one data-set, while the scores on the two other data-sets remain unchanged.
15
Haji, Sismat Muhamad Alif Bin. "Quality and productivity : a comparative analysis of human translation and post-editing with Malay learners of Arabic and English." Thesis, University of Leeds, 2016. http://etheses.whiterose.ac.uk/17549/.
Full textAbstract:
Translation into and between foreign languages has become a common practice in the professional setting. However, this translation directionality has yet to be thoroughly explored, especially when post-editing is involved. The present study conducts experiments on the application of machine translation (MT) and translation memory (TM) in a translation classroom setting. A group of Malay speakers, who are non-native speakers of Arabic and English, used MemoQ 2014 to translate technical Arabic and English texts by post-editing raw MT and modified TM outputs containing several errors. The non-native trainee translators’ productivity was measured and the quality of the translation was assessed through error analysis approach based on the MeLLANGE error typology so that it could provide a comprehensive analysis of the types of errors commonly found in the non-native trainee translators’ translations. The error annotation also aims to provide guidelines for translators who work with the Arabic-English language pair and non-native translators. The present study revealed that the translation technologies helped improve the non-native translators’ speed and quality. The study also discovered that syntactic and lexical errors are the most problematic in the PE tasks. The trainee translators tend to overlook the errors that were caused by cross-linguistic influence, such as articles, gender, number and the conjunction “wa”. However, this could have been avoided if the participants revised their translations thoroughly because most of the errors are minor. The study also revealed that the non-native trainee translators could be as productive as the professional native translators because they managed to reach the average daily productivity for professional translators, which is at least 5,000 words per day.
16
Bacusmo, Jo Marie. "Proline Codon Translational Fidelity in Rhodopseudomonas palustris: Characterization of Novel Trans-editing Factor ProXp-abu." The Ohio State University, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=osu1397747141.
Full text
17
Huang, Jin. "Working styles of student translators in revision and post-editing : an empirical-experimental study with eye-tracking, keylogging and cue-based retrospection." Thesis, Durham University, 2016. http://etheses.dur.ac.uk/11594/.
Full textAbstract:
In today’s translation profession, being skilful at revision (including self-revision and other-revision) and post-editing tasks is becoming essential for translators. The exploration of the working styles of student translators in the revision and post-editing processes is vital in helping us to understand the nature of these tasks, and may help in improving pedagogy. Drawing on theories from translation-related studies, cognitive psychology, and text comprehension and production, the aims of this research were to: (1) identify the basic types of reading and typing activity (physical activities) of student translators in the processes of revision and post-editing, and to measure statistically and compare the duration of these activities within and across tasks; (2) identify the underlying purposes (mental activities) behind each type of reading and typing activity; (3) categorise the basic types of working style of student translators and compare the frequency of use of each working style both within and across tasks; (4) identify the personal working styles of student translators in carrying out different tasks, and (5) identify the most efficient working style in each task. Eighteen student translators from Durham University, with Chinese as L1 and English as L2, were invited to participate in the experiment. They were asked to translate, self-revise, other-revise and post-edit three comparable texts in Translog-II with the eye-tracking plugin activated. A cue-based retrospective interview was carried out after each session to collect the student translators’ subjective and conscious data for qualitative analysis. The raw logging data were transformed into User Activity Data and were analysed both quantitatively and qualitatively. This study identified seven types of reading and typing activity in the processes of self-revision, other-revision and post-editing. Three revision phases were defined and four types of working style were recognised. The student translators’ personal working styles were compared in all three tasks. In addition, a tentative model of their cognitive processes in self-revision, other-revision and post-editing was developed, and the efficiency of the four working styles in each task was tested.
18
Cervi, Stefano. "Traduzione assistita, manuale e post-editing: Un progetto di analisi comparativa per la traduzione in italiano di ricette di cucina in lingua francese." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2014. http://amslaurea.unibo.it/7453/.
Full textAbstract:
Nel mondo della traduzione non sempre si può guardare esclusivamente alla qualità del testo di arrivo: ci sono scenari traduttivi, come quello oggetto del presente elaborato, in cui è necessario tener conto di altri fattori, in primis i costi per il committente e la produttività del traduttore.
Con questo elaborato intendo dimostrare che per lo scenario traduttivo preso in esame, ossia la traduzione per un sito web di una grande quantità di ricette da parte di traduttori diversi, l’ausilio di un programma di traduzione assistita è da preferire, per produttività, coerenza traduttiva e contenimento dei costi, alla traduzione manuale e al post-editing della traduzione automatica.
Per tale scopo, ho tradotto una ricetta con ciascuna di queste metodologie di lavoro, così da poterle mettere a confronto e potermi pronunciare, a seguito di un'analisi approfondita, circa il metodo migliore per lo scenario descritto.
19
Blain, Frédéric. "Modèles de traduction évolutifs." Thesis, Le Mans, 2013. http://www.theses.fr/2013LEMA1034/document.
Full textAbstract:
Bien que la recherche ait fait progresser la traduction automatique depuis plusieurs années, la sortie d’un système automatisé ne peut être généralement publiée sans avoir été révisée humainement au préalable, et corrigée le cas échéant. Forts de ce constat, nous avons voulu exploiter ces retours utilisateurs issus du processus de révision pour adapter notre système statistique dans le temps, au moyen d’une approche incrémentale.Dans le cadre de cette thèse Cifre-Défense, nous nous sommes donc intéressés à la postédition, un des champs de recherche les plus actifs du moment, et qui plus est très utilisé dans l’industrie de la traduction et de la localisation.L’intégration de retours utilisateurs n’est toutefois pas une tâche aussi évidente qu’il n’y paraît. D’une part, il faut être capable d’identifier l’information qui sera utile au système, parmi l’ensemble des modifications apportées par l’utilisateur. Pour répondre à cette problématique, nous avons introduit une nouvelle notion (les « Actions de Post-Édition »), et proposé une méthodologie d’analyse permettant l’identification automatique de cette information à partir de données post-éditées. D’autre part, concernant l’intégration continue des retours utilisateurs nous avons développé un algorithme d’adaptation incrémentale pour un système de traduction statistique, lequel obtient des performances supérieures à la procédure standard. Ceci est d’autant plus intéressant que le développement et l’optimisation d’un tel système de traduction estune tâche très coûteuse en ressources computationnelles, nécessitant parfois jusqu’à plusieurs jours de calcul.Conduits conjointement au sein de l’entreprise SYSTRAN et du LIUM, les travaux de recherche de cette thèse s’inscrivent dans le cadre du projet ANR COSMAT 1. En partenariat avec l’INRIA, ce projet avait pour objectif de fournir à la communauté scientifique un service collaboratif de traduction automatique de contenus scientifiques. Outre les problématiques liéesà ce type de contenu (adaptation au domaine, reconnaissance d’entités scientifiques, etc.), c’est l’aspect collaboratif de ce service avec la possibilité donnée aux utilisateurs de réviser les traductions qui donne un cadre applicatif à nos travaux de rechercheAlthough machine translation research achieved big progress for several years, the output of an automated system cannot be published without prior revision by human annotators. Based on this fact, we wanted to exploit the user feedbacks from the review process in order to incrementally adapt our statistical system over time.As part of this thesis, we are therefore interested in the post-editing, one of the most active fields of research, and what is more widely used in the translation and localization industry.However, the integration of user feedbacks is not an obvious task. On the one hand, we must be able to identify the information that will be useful for the system, among all changes made by the user. To address this problem, we introduced a new concept (the “Post-Editing Actions”), and proposed an analysis methodology for automatic identification of this information from post-edited data. On the other hand, for the continuous integration of user feedbacks, we havedeveloped an algorithm for incremental adaptation of a statistical machine translation system, which gets higher performance than the standard procedure. This is even more interesting as both development and optimization of this type of translation system has a very computational cost, sometimes requiring several days of computing.Conducted jointly with SYSTRAN and LIUM, the research work of this thesis is part of the French Government Research Agency project COSMAT 2. This project aimed to provide a collaborative machine translation service for scientific content to the scientific community. The collaborative aspect of this service with the possibility for users to review the translations givesan application framework for our research
20
Brizzi, Mattia. "La traduzione automatica dall’italiano in francese e il linguaggio dell’arte: post-editing dell’output di DeepL Traduttore a partire dal sito di Piero della Francesca." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/20403/.
Full textAbstract:
Abstract
La presente tesi illustra il processo di post-editing completo dell’output di DeepL Traduttore a partire dalle sezioni “Biografia” e “Le Opere” del sito http://www.pierodellafrancesca.it/. Il primo capitolo traccia un quadro della storia della traduzione automatica e del ruolo che questa riveste al giorno d’oggi. In relazione a tale aspetto, viene fornita una definizione del concetto di post-editing e vengono illustrate le modalità e gli scenari in cui tale attività può essere svolta. Il secondo capitolo descrive la metodologia di lavoro adottata per la revisione dell’output di DeepL, con un approfondimento sul tema della traduzione dei nomi propri. In seguito, si fornisce una descrizione dei principali strumenti di valutazione della qualità dei sistemi di TA e si espone la tassonomia con cui sono state successivamente classificate le revisioni effettuate. Tale tassonomia prende ampio spunto da quelle elaborate in Temnikova (2010) e Koponen et al. (2012). Il terzo capitolo è dedicato all’analisi del testo di partenza, che si articola in parallelo rispetto al lavoro di editing e si suddivide in tre livelli: lessicale, morfo-sintattico e stilistico. Il quarto capitolo presenta le proposte di revisione avanzate per portare la qualità l’output grezzo al livello di quella di un testo pubblicabile. Infine, nel quinto capitolo, si ricorre alla tassonomia adottata per osservare la distribuzione delle correzioni apportate. L’osservazione delle tipologie di interventi di editing effettuati, intrinsecamente legati agli errori commessi dal traduttore automatico, è funzionale all’elaborazione di un giudizio complessivo sulla qualità dell’output fornito da DeepL Traduttore.
21
McMillan, Matthew Christopher. "Editing the independent, digital feature film, mosaic." Thesis, Queensland University of Technology, 2007. https://eprints.qut.edu.au/16558/1/Matt_McMillan_Thesis.pdf.
Full textAbstract:
The production of the independent, digital feature film titled Mosaic was performed on a very low budget. The design and implementation of the post-production of the film required consideration of budgetary constraints, and solutions to these constraints that would still allow the creative freedom of the editor and the director.
The technical design was based around digital filmmaking technology. The choice of this technology influenced how the editor was able to address aesthetic and technical challenges.
22
McMillan, Matthew Christopher. "Editing the independent, digital feature film, mosaic." Queensland University of Technology, 2007. http://eprints.qut.edu.au/16558/.
Full textAbstract:
The production of the independent, digital feature film titled Mosaic was performed on a very low budget. The design and implementation of the post-production of the film required consideration of budgetary constraints, and solutions to these constraints that would still allow the creative freedom of the editor and the director. The technical design was based around digital filmmaking technology. The choice of this technology influenced how the editor was able to address aesthetic and technical challenges.
23
Rémillard, Judith. "Utilité et utilisation de la traduction automatique dans l’environnement de traduction : une évaluation axée sur les traducteurs professionnels." Thesis, Université d'Ottawa / University of Ottawa, 2018. http://hdl.handle.net/10393/37784.
Full textAbstract:
L’arrivée de la traduction automatique (TA) bouleverse les pratiques dans l’industrie de la traduction et soulève par le fait même des questions sur l’utilité et l’utilisation de cette technologie. Puisque de nombreuses études ont déjà porté sur son utilisation dans un contexte où elle est imposée aux traducteurs, nous avons choisi d’adopter la perspective toute particulière des traducteurs pour examiner son utilité et son utilisation volontaire dans l’environnement de traduction (ET). Notre recherche visait à répondre à trois grandes questions : les traducteurs utilisent-ils la TA dans leurs pratiques? Les traducteurs croient-ils que les données de sortie sont utiles et utilisables? Les traducteurs utilisent-ils concrètement ces données de sortie dans le processus de traduction?
Pour répondre à ces questions, nous avons d’abord diffusé un sondage à grande échelle afin de mesurer l’utilisation de la TA en tant qu’outil, de recueillir des données sur le profil des répondants et d’évaluer leur perception d’utilité par rapport aux données de sortie et aux divers types de phénomènes que nous avions identifiés au préalable avec l’aide de deux traducteurs professionnels. Ensuite, nous avons réalisé une expérience avec d’autres traducteurs professionnels où nous leur avons demandé de procéder à la traduction de courts segments et avons examiné s’ils utilisaient ou non ces données de sortie pour produire leur traduction. Notre analyse était fondée sur le principe que, dans un contexte d’utilisation volontaire, l’utilisation d’une donnée de sortie permet d’induire une perception d’utilité et d’examiner, par le fait même, l’utilité de la TA. Dans l’ensemble, nous avons trouvé que la TA n’est habituellement pas utilisée de façon volontaire, que les perceptions des traducteurs sont peu favorables à une telle utilisation, que la perception des traducteurs quant à l’utilité des données de sortie est aussi plutôt négative, mais que les données de sortie semblent être beaucoup plus utiles et utilisables que ce que ne pourraient le croire les traducteurs, car ils les ont généralement utilisées dans le processus de traduction. Nous avons aussi examiné les facteurs déterminants de l’utilité et de l’utilisation de la TA et des données de sortie.
24
Stevens, Grant William. "The space of editing : playing with difference in art, film and writing." Thesis, Queensland University of Technology, 2007. https://eprints.qut.edu.au/16361/1/Grant_Stevens_Thesis.pdf.
Full textAbstract:
This research project explores the creative and critical functions of editing in art, film and writing. The written component analyses the histories and discourses of 'cutting and splicing' to examine their various roles in processes of signification. The artistic practice uses more speculative and open-ended methods to explore the social 'languages' that inform our inter-subjective experiences. This project argues that editing is a creative methodology for making meaning, because it allows existing symbolic systems to be appropriated, revised and rewritten. By emphasising the operations of spacing, questioning and play, it also identifies editing as an essential tool for critically engaging with the potentials of art and theory.
25
Stevens, Grant William. "The space of editing : playing with difference in art, film and writing." Queensland University of Technology, 2007. http://eprints.qut.edu.au/16361/.
Full textAbstract:
This research project explores the creative and critical functions of editing in art, film and writing. The written component analyses the histories and discourses of 'cutting and splicing' to examine their various roles in processes of signification. The artistic practice uses more speculative and open-ended methods to explore the social 'languages' that inform our inter-subjective experiences. This project argues that editing is a creative methodology for making meaning, because it allows existing symbolic systems to be appropriated, revised and rewritten. By emphasising the operations of spacing, questioning and play, it also identifies editing as an essential tool for critically engaging with the potentials of art and theory.
26
Busi, Gioia. "Changes in the translation industry: A prospectus on the near future of innovation in machine translation." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2019.
Find full textAbstract:
This thesis aims to analyze the supposed inevitability of a breakthrough in machine translation and the role that this breakthrough will play in the evolution of translation companies. It will analyze the changes that are happening and what repercussions those changes will have on the decisions made by students, professionals, agencies, and institutions over the next twenty years.
This paper will be divided into three main sections:
The first part will provide a background of today’s translation industry and consider the advent of machine translation in translation agencies and its continuous developments.
In the second part of this essay, I will illustrate how I carried out my research inside Global Voices - a translation agency based in Stirling, Scotland, in which I have interned as Project Manager in December 2018 - to understand what use the translation agency makes of machine translation, also conveying my colleagues’ thoughts about it.
The conclusion will recapitulate the topics approached, revolve around the main findings of this study and try to foresee what translators should expect from the future, how in my opinion they should deal with the changes the future will bring.
27
Mambelli, Agnese. "I limiti della traduzione automatica nel linguaggio del turismo e l'importanza del traduttore umano: il caso del sito web Turismo de Cantabria." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2021.
Find full textAbstract:
L’obiettivo del presente elaborato è quello di sottolineare i limiti della traduzione automatica, in particolar modo legati al linguaggio del turismo, e di riattribuire importanza al ruolo attivo del traduttore umano. A tale scopo, si presenta sia un’analisi della traduzione di Google Traduttore di estratti del sito web Turismo de Cantabria sia una proposta personale di traduzione a seguito di un intervento di post-editing. Il primo capitolo fornisce cenni teorici sull’evoluzione della traduzione automatica, sui sistemi di traduzione e sulle difficoltà riscontrate se ne si fa un uso esclusivo, soprattutto nel momento in cui si affrontano tipologie testuali e linguaggi specifici come quello del turismo; per ultimo, si introduce brevemente la nuova figura del post-editor che si inserisce nel mondo della traduzione per valorizzare il lavoro dell’uomo. Il secondo capitolo, invece, prende in esame le due versioni della traduzione automatica e propone una versione post-editata. Infine, il terzo capitolo è dedicato all’analisi dei problemi di traduzione del sistema di TA, degli errori maggiormente riscontrati nelle versioni di Google e delle tecniche traduttive implementate nel post-editing.
28
Mendoza, García María Helena. "La posedición de traducciones de textos técnicos del alemán al castellano." Doctoral thesis, Universitat Autònoma de Barcelona, 2017. http://hdl.handle.net/10803/457362.
Full textAbstract:
Esta tesis pretende realizar un estudio descriptivo de la problemática de la traducción de textos técnicos del alemán al castellano mediante traducción automática y elaborar una guía de ayuda para la posedición de dichas traducciones.
La hipótesis parte del hecho de que el resultado de las traducciones automáticas hasta el momento no es satisfactorio; que presentan carencias que sólo se pueden solucionar mediante el proceso de la posedición. Nuestra hipótesis consiste en que es posible sistematizar el tipo de problemas de traducción provocados por el uso de la traducción automática en una combinación lingüística y un género determinados y que la identificación de estos problemas permitirá elaborar una guía de directrices para la posedición.
Traductología, traducción automática, tecnologías de la traducción, posedición, errores en la traducción, problemas de la traducción, directrices para la posedición, motores de traducción automática, control de calidad de la traducción, formación del poseditor, ayuda en la posedición.At present, automatic translations of technical texts from one language to another show deficiencies that can only be solved through a post-editing process. This Thesis intends to study the problems encountered in the machine translation of technical texts from German into Spanish and create a practical guide that helps with the post-editing of those translations. We hypothesize that it is possible to implement a system to identify the problems found in an automatic translation in a linguistic combination and a settled genre that would allow us to create a coherent, established guide to help with the post-editing process.
29
Marie, Benjamin. "Exploitation d’informations riches pour guider la traduction automatique statistique." Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLS066/document.
Full textAbstract:
S'il est indéniable que de nos jours la traduction automatique (TA) facilite la communication entre langues, et plus encore depuis les récents progrès des systèmes de TA statistiques, ses résultats sont encore loin du niveau de qualité des traductions obtenues avec des traducteurs humains.Ce constat résulte en partie du mode de fonctionnement d'un système de TA statistique, très contraint sur la nature des modèles qu'il peut utiliser pour construire et évaluer de nombreuses hypothèses de traduction partielles avant de parvenir à une hypothèse de traduction complète. Il existe cependant des types de modèles, que nous qualifions de « complexes », qui sont appris à partir d'informations riches. Si un enjeu pour les développeurs de systèmes de TA consiste à les intégrer lors de la construction initiale des hypothèses de traduction, cela n'est pas toujours possible, car elles peuvent notamment nécessiter des hypothèses complètes ou impliquer un coût de calcul très important. En conséquence, de tels modèles complexes sont typiquement uniquement utilisés en TA pour effectuer le reclassement de listes de meilleures hypothèses complètes. Bien que ceci permette dans les faits de tirer profit d'une meilleure modélisation de certains aspects des traductions, cette approche reste par nature limitée : en effet, les listes d'hypothèses reclassées ne représentent qu'une infime partie de l'espace de recherche du décodeur, contiennent des hypothèses peu diversifiées, et ont été obtenues à l'aide de modèles dont la nature peut être très différente des modèles complexes utilisés en reclassement.Nous formulons donc l'hypothèse que de telles listes d'hypothèses de traduction sont mal adaptées afin de faire s'exprimer au mieux les modèles complexes utilisés. Les travaux que nous présentons dans cette thèse ont pour objectif de permettre une meilleure exploitation d'informations riches pour l'amélioration des traductions obtenues à l'aide de systèmes de TA statistique.Notre première contribution s'articule autour d'un système de réécriture guidé par des informations riches. Des réécritures successives, appliquées aux meilleures hypothèses de traduction obtenues avec un système de reclassement ayant accès aux mêmes informations riches, permettent à notre système d'améliorer la qualité de la traduction.L'originalité de notre seconde contribution consiste à faire une construction de listes d'hypothèses par passes multiples qui exploitent des informations dérivées de l'évaluation des hypothèses de traduction produites antérieurement à l'aide de notre ensemble d'informations riches. Notre système produit ainsi des listes d'hypothèses plus diversifiées et de meilleure qualité, qui s'avèrent donc plus intéressantes pour un reclassement fondé sur des informations riches. De surcroît, notre système de réécriture précédent permet d'améliorer les hypothèses produites par cette deuxième approche à passes multiples.Notre troisième contribution repose sur la simulation d'un type d'information idéalisé parfait qui permet de déterminer quelles parties d'une hypothèse de traduction sont correctes. Cette idéalisation nous permet d'apporter une indication de la meilleure performance atteignable avec les approches introduites précédemment si les informations riches disponibles décrivaient parfaitement ce qui constitue une bonne traduction. Cette approche est en outre présentée sous la forme d'une traduction interactive, baptisée « pré-post-édition », qui serait réduite à sa forme la plus simple : un système de TA statistique produit sa meilleure hypothèse de traduction, puis un humain apporte la connaissance des parties qui sont correctes, et cette information est exploitée au cours d'une nouvelle recherche pour identifier une meilleure traductionAlthough communication between languages has without question been made easier thanks to Machine Translation (MT), especially given the recent advances in statistical MT systems, the quality of the translations produced by MT systems is still well below the translation quality that can be obtained through human translation. This gap is partly due to the way in which statistical MT systems operate; the types of models that can be used are limited because of the need to construct and evaluate a great number of partial hypotheses to produce a complete translation hypothesis. While more “complex” models learnt from richer information do exist, in practice, their integration into the system is not always possible, would necessitate a complete hypothesis to be computed or would be too computationally expensive. Such features are therefore typically used in a reranking step applied to the list of the best complete hypotheses produced by the MT system.Using these features in a reranking framework does often provide a better modelization of certain aspects of the translation. However, this approach is inherently limited: reranked hypothesis lists represent only a small portion of the decoder's search space, tend to contain hypotheses that vary little between each other and which were obtained with features that may be very different from the complex features to be used during reranking.In this work, we put forward the hypothesis that such translation hypothesis lists are poorly adapted for exploiting the full potential of complex features. The aim of this thesis is to establish new and better methods of exploiting such features to improve translations produced by statistical MT systems.Our first contribution is a rewriting system guided by complex features. Sequences of rewriting operations, applied to hypotheses obtained by a reranking framework that uses the same features, allow us to obtain a substantial improvement in translation quality.The originality of our second contribution lies in the construction of hypothesis lists with a multi-pass decoding that exploits information derived from the evaluation of previously translated hypotheses, using a set of complex features. Our system is therefore capable of producing more diverse hypothesis lists, which are globally of a better quality and which are better adapted to a reranking step with complex features. What is more, our forementioned rewriting system enables us to further improve the hypotheses produced with our multi-pass decoding approach.Our third contribution is based on the simulation of an ideal information type, designed to perfectly identify the correct fragments of a translation hypothesis. This perfect information gives us an indication of the best attainable performance with the systems described in our first two contributions, in the case where the complex features are able to modelize the translation perfectly. Through this approach, we also introduce a novel form of interactive translation, coined "pre-post-editing", under a very simplified form: a statistical MT system produces its best translation hypothesis, then a human indicates which fragments of the hypothesis are correct, and this new information is then used during a new decoding pass to find a new best translation
30
Bystrova-McIntyre, Tatyana. "Cohesion in Translation: A Corpus Study of Human-translated, Machine-translated, and Non-translated Texts (Russian into English)." Kent State University / OhioLINK, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=kent1353451112.
Full text
31
Zhang, Hong. "Formación en posedición de traducción automática para estudiantes de lenguas extranjeras." Doctoral thesis, Universitat Autònoma de Barcelona, 2019. http://hdl.handle.net/10803/669645.
Full textAbstract:
La presente tesis doctoral trata sobre la formación en posedición (PE) de traducción automática (TA) aplicada a la didáctica del español para estudiantes chinos. El objetivo general de esta tesis es explorar si la PE puede ser una herramienta útil en la enseñanza del español como segunda lengua para los estudiantes chinos con un nivel mínimo de B2. Más concretamente, investigaremos si una propuesta didáctica elaborada en esta tesis resulta útil y válida para ayudar a los estudiantes a identificar y corregir errores producidos mediante TA. A pesar de que los profesores de idiomas extranjeros normalmente no recomiendan el uso de TA, lo cierto es que muchos estudiantes emplean este recurso de traducción.
Hemos realizado dos estudios con dos grupos de estudiantes chinos distintos. Hemos utilizado la grabación de pantalla para registrar todas las pruebas que hemos hecho y tomamos nota de cada acción realizada por los estudiantes. El primer estudio consistió en dos partes y se realizó en el año 2017. La primera parte consistía en una prueba de traducción corta para ver si los estudiantes podían traducir mejor con o sin la asistencia de TA. En la segunda parte preparamos un texto traducido mediante un motor de traducción automática y propusimos a los estudiantes chinos identificar y corregir los errores de la traducción generada por TA. A partir de estos resultados se elaboraron materiales de formación en PE. El segundo estudio se realizó en 2018. Durante 60 minutos, en el pretest, los estudiantes tuvieron que poseditar un texto traducido automáticamente. Después del pretest, los estudiantes se dividieron en dos grupos: grupo experimental y grupo de control. El grupo experimental realizó la formación en PE y el grupo de control no. En el postest, ambos grupos realizaron PE del mismo texto del pretest.
Al comparar los resultados de dos grupos, el grupo experimental pudo identificar y corregir más errores que el grupo de control en el postest. El grupo experimental obtuvo mejores resultados especialmente en omisión, orden de palabras y nombre oficial después de la formación en PE. Concretamente, poseditó mejor y en menos tiempo que el grupo de control en el postest.
A partir de los resultados de este estudio, podemos ver que la formación en PE puede ayudar a los estudiantes chinos a identificar y corregir los errores de omisión, orden de palabras y nombre oficial, pero parece que durante el breve tiempo del curso, los errores de precisión, preposición y el registro formal fueron difíciles para los estudiantes.This doctoral thesis deals machine translation (MT) post-editing (PE) training applied to the teaching of Spanish for Chinese students. Although the use of MT is not usually recommended by foreign language teachers, many students use MT regardless; therefore, the general objective of this thesis is to explore whether PE can be a useful tool in teaching Spanish as a second language to Chinese students with a minimum level of B2. More specifically, we will investigate whether the didactic proposal elaborated in this thesis is a useful and valid method to help students identify and correct errors produced by raw output of MT. We performed two experiments with two different groups of Chinese students. We used screen recording to record all the tests we administered, and made notes of every action performed by the students. The first experiment, which conducted in 2017, included two parts: the first part consisted of a short translation test to see whether or not Chinese students could write better Spanish with the assistance of MT. In the second part, we prepared a translated text using a machine translation engine and asked Chinese students to identify and correct the mistakes of the of the MT output. Based on these results, PE training materials were developed. The second study was conducted in 2018. For 60 minutes, in the pretest, the students had to post-edit an automatically translated text. After the pretest, the students were divided into two groups: the experimental group and the control group. The experimental group participated in the PE training and the control group did not. In the post-test, both groups performed the PE of the same pretest text. By comparing the results of two groups, it was evident that the experimental group could identify and correct more errors than the control group in the post-test. The experimental group did significantly better in identifying errors in omission, word order, and official name after the PE training; they produced more accurate edits and required less time than control group in the post-test. From the results of this study, we can see that PE training can help Chinese students to identify and correct errors of omission, word order, and official name in a raw MT output from Chinese to Spanish; however, during this short-term course, they still had difficulties with accuracy, prepositions, and formal register.
32
Temizöz, Özlem. "Postediting machine translation output and its revision: subject-matter experts versus professional translators." Doctoral thesis, Universitat Rovira i Virgili, 2013. http://hdl.handle.net/10803/128204.
Full textAbstract:
El presente estudio compara la post-edición de textos técnicos de ingenieros y traductores profesionales en términos de velocidad, documentación y cambios. También se compara la calidad de los textos post-editados. Además, se explora cuál de los siguientes flujos de trabajo es más rápido y produce resultados de mayor calidad: la post-edición de los resultados de Traducción Automática hecha por los ingenieros y la revisada por traductores profesionales, o viceversa. Los resultados sugieren que la experiencia y conocimientos en la materia son los principales factores que determinan la calidad de la post-edición. Cuando se penalizan los errores recurrentes, la post-edición de textos técnicos realizada por los ingenieros es significativamente de mayor calidad que la de los traductores. La velocidad de revisión de traductores e ingenieros no difirió significativamente. En textos técnicos, la mejora de la calidad que conlleva que el ingeniero revise la post-edición del traductor es mayor que en cuando el trabajo se organiza al revés. Además, la calidad de los textos post-editados y sus versiones revisadas (ya sea realizada por traductores profesionales o ingenieros) cambia significativamente según se penalicen o no los errores recurrentes.El present estudi compara la post-edició de textos tècnics d'enginyers i traductors professionals en termes de velocitat, documentació i canvis. També es compara la qualitat dels textos post-editats. A més, s'explora quin dels següents fluxos de treball és més ràpid i produeix resultats de major qualitat: la post-edició dels resultats de Traducció Automàtica feta pels enginyers i la revisada per traductors professionals, o viceversa. Els resultats suggereixen que l'experiència i coneixements en la matèria són els principals factors que determinen la qualitat de la post-edició. Quan es penalitzen els errors recurrents, la post-edició de textos tècnics realitzada pels enginyers és significativament de major qualitat que la dels traductors. La velocitat de revisió de traductors i enginyers no va diferir significativament. En textos tècnics, la millora de la qualitat que comporta que l'enginyer revisi la post-edició del traductor és major que en quan el treball s'organitza a l'inrevés. A més, la qualitat dels textos post-editats i les seves versions revisades (ja sigui realitzada per traductors professionals o enginyers) canvia significativament segons es penalitzin o no els errors recurrents.
The present research compares engineers’ and professional translators’ postediting a technical text in terms of speed, documentation and changes. It also compares the postedited texts with regard to quality. Further, we explore which of the following workflows is faster and produces outputs of higher quality: Postediting MT output by engineers and revising the postedited text by professional translators, or vice-versa. The findings suggest that expertise and experience in the subject-matter are the main factors determining postediting quality. When the recurrent errors are penalized, the engineers’ postediting of technical texts is of significantly higher quality than the translators’. The translators’ and the engineers’ postediting and revision speed did not differ significantly. For technical texts, the quality improvement brought about by engineer-revision of translator-postediting is higher than vice-versa. Further, the quality of the postedited texts and their revised versions (either performed by professional translators or engineers) changes significantly as a result of penalizing and unpenalizing recurrent errors.
33
Pagin, Elia. "La traduzione a servizio dell’internazionalizzazione d’impresa: l’output del programma di traduzione automatica DeepL a confronto con i risultati della traduzione assistita." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2018. http://amslaurea.unibo.it/15269/.
Full textAbstract:
Il traduttore riveste un ruolo chiave nel processo di internazionalizzazione delle imprese, non solo garantendo la veicolazione dei contenuti aziendali in altre lingue, ma anche sensibilizzando le imprese stesse a una fruizione di servizi linguistici di qualità. Da queste considerazioni trae le proprie basi il progetto Language Toolkit, nel cui ambito è stato redatto il presente elaborato. Trattasi della traduzione dall’italiano verso il tedesco di un manuale tecnico per l’azienda ATI di Cesena, dapprima eseguita con l’ausilio di uno strumento di traduzione assistita e di seguito posta a confronto con l’output generato dal sistema di traduzione automatica DeepL, di recente introduzione.
L’obiettivo del presente elaborato consta nella valutazione della qualità dell’output grezzo di DeepL e della sua applicabilità alla traduzione di testi specialistici, in vista di un risparmio di risorse e di tempo. L’analisi è stata condotta ricorrendo a due tipologie di valutazione, la prima di tipo automatico mediante l’algoritmo BLEU score, e la seconda svolta dallo stesso traduttore umano sulla base degli interventi di post-editing necessari affinché la traduzione possa essere pubblicata.
34
Mellinger, Christopher Davey. "Computer-Assisted Translation: An Empirical Investigation of Cognitive Effort." Kent State University / OhioLINK, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=kent1398088162.
Full text
35
Santi, Greta. "La nuova frontiera della traduzione: la localizzazione di un sito web." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2020.
Find full textAbstract:
L’obiettivo dell’elaborato è favorire l’ampliamento del mercato dell’azienda agricola Rio del Sol nei paesi germanofoni. In particolare, viene tradotto in lingua tedesca il sito web aziendale, estremamente importante per la presentazione su scala internazionale dei prodotti e per lo sviluppo dell’e-commerce, risorsa fondamentale, specialmente in questo periodo segnato dall’emergenza da Coronavirus. Infatti, ora più che mai, molti preferiscono fare acquisti comodamente da casa. Quindi, la localizzazione di siti web, destinata a marcare il nostro futuro, rappresenta la nuova frontiera della traduzione. La localizzazione del sito web Rio del Sol viene fatta con l’aiuto di DeepL, strumento di traduzione automatica, del quale si cerca di valutare l’efficacia. In più, tale lavoro comporta molteplici competenze e non si limita alla mera traduzione di alcuni testi: un sito web, come quello di Rio del Sol, si articola spesso in numerose voci di menu contenenti link extra-testuali ad articoli di giornale o testi audiovisivi. Infatti, l’elaborato affronta anche la localizzazione di un articolo apparso su Corriere Romagna e di un’intervista alla titolare dell’azienda pubblicata su YouTube da Italia nel Bicchiere. Il primo capitolo dell’elaborato tratta, perciò, delle caratteristiche del sito web Rio del Sol e delle particolarità della localizzazione del sito stesso e dell’articolo di giornale. Viene inoltre descritto l’utilizzo dei programmi OmegaT e SDL Trados Studio, necessari per la creazione di documenti in lingua tedesca. Il secondo capitolo si concentra, invece, sulla sottotitolazione del video pubblicato su YouTube, descrivendone norme e realizzazione tramite Subtitle Edit. Infine, il terzo capitolo analizza il lavoro svolto da DeepL, valutando se si tratti di un basilare software per la traduzione automatica o di un efficiente supporto al post-editing. Per trarre conclusioni valide vengono utilizzate pubblicazioni accademiche e l’esperienza maturata in corso d’opera.
36
Martins, Débora Beatriz de Jesus. "Pós-edição automática de textos traduzidos automaticamente de inglês para português do Brasil." Universidade Federal de São Carlos, 2014. https://repositorio.ufscar.br/handle/ufscar/563.
Full textAbstract:
Made available in DSpace on 2016-06-02T19:06:12Z (GMT). No. of bitstreams: 1
5932.pdf: 1110060 bytes, checksum: fe08b552e37f04451248c376cfc4454f (MD5)
Previous issue date: 2014-04-10Universidade Federal de Minas Gerais
The project described in this document focusses on the post-editing of automatically translated texts. Machine Translation (MT) is the task of translating texts in natural language performed by a computer and it is part of the Natural Language Processing (NLP) research field, linked to the Artificial Intelligence (AI) area. Researches in MT using different approaches, such as linguistics and statistics, have advanced greatly since its beginning in the 1950 s. Nonetheless, the automatically translated texts, except when used to provide a basic understanding of a text, still need to go through post-editing to become well written in the target language. At present, the most common form of post-editing is that executed by human translators, whether they are professional translators or the users of the MT system themselves. Manual post-editing is more accurate but it is cost and time demanding and can be prohibitive when too many changes have to be made. As an attempt to advance in the state-of-the-art in MT research, mainly regarding Brazilian Portuguese, this research has as its goal verifying the effectiveness of using an Automated Post-Editing (APE) system in translations from English to Portuguese. By using a training corpus containing reference translations (good translations produced by humans) and translations produced by a phrase-based statistical MT system, machine learning techniques were applied for the APE creation. The resulting APE system is able to: (i) automatically identify MT errors and (ii) automatically correct MT errors by using previous error identification or not. The evaluation of the APE effectiveness was made through the usage of the automatic evaluation metrics BLEU and NIST, calculated for post-edited and not post-edited sentences. There was also manual verification of the sentences. Despite the limited results that were achieved due to the small size of our training corpus, we can conclude that the resulting APE improves MT quality from English to Portuguese.
O projeto de mestrado descrito neste documento tem como foco a pós-edição de textos traduzidos automaticamente. Tradução Automática (TA) é a tarefa de traduzir textos em língua natural desempenhada por um computador e faz parte da linha de pesquisa de Processamento de Línguas Naturais (PLN), vinculada à área de Inteligência Artificial (IA). As pesquisas em TA, utilizando desde abordagens linguísticas até modelos estatísticos, têm avançado muito desde seu início na década de 1950. Entretanto, os textos traduzidos automaticamente, exceto quando utilizados apenas para um entendimento geral do assunto, ainda precisam passar por pós-edição para que se tornem bem escritos na língua alvo. Atualmente, a forma mais comum de pós-edição é a executada por tradutores humanos, sejam eles profissionais ou os próprios usuários dos sistemas de TA. A pós-edição manual é mais precisa, mas traz custo e demanda tempo, especialmente quando envolve muitas alterações. Como uma tentativa para avançar o estado da arte das pesquisas em TA, principalmente envolvendo o português do Brasil, esta pesquisa visa verificar a efetividade do uso de um sistema de pós-edição automática (Automated Post-Editing ou APE) na tradução do inglês para o português. Utilizando um corpus de treinamento contendo traduções de referência (boas traduções produzidas por humanos) e traduções geradas por um sistema de TA estatística baseada em frases, técnicas de aprendizado de máquina foram aplicadas para o desenvolvimento do APE. O sistema de APE desenvolvido: (i) identifica automaticamente os erros de TA e (ii) realiza a correção automática da tradução com ou sem a identificação prévia dos erros. A avaliação foi realizada usando tanto medidas automáticas BLEU e NIST, calculadas para as sentenças sem e com a pós-edição; como analise manual. Apesar de resultados limitados pelo pequeno tamanho do corpus de treinamento, foi possível concluir que o APE desenvolvido melhora a qualidade da TA de inglês para português.
37
Nozaic, Claire. "An introduction to audio post-production for film." Thesis, Stellenbosch : Stellenbosch University, 2006. http://hdl.handle.net/10019.1/17405.
Full textAbstract:
Thesis (M.Mus.)--University of Stellenbosch, 2006.ENGLISH ABSTRACT: In South Africa there has been an increase over the last few years in audio engineering courses which include modules of study in audio post-production or even offer audio post-production as a major focus of study. From an academic standpoint however, and despite the growth in the local film industry, very little study of this field has been undertaken in South Africa until recently. In 2005, a MMus thesis was submitted at the University of KwaZulu-Natal entitled Acoustic Ambience in Cinematography: An Exploration of the Descriptive and Emotional Impact of the Aural Environment (Turner, 2005: online). The thesis briefly outlines the basic components of the soundtrack and focuses on describing and analysing the properties of ambience, a sub-section of sound effects. At Stellenbosch University, research has recently begun in the fields of film music and Foley (sound effects associated with human movement onscreen). The purpose of this thesis is to provide an overview of audio post-production and the contribution of sound to the film medium. It provides an outline of the processes involved in creating a soundtrack for film and includes a description of the components of the soundtrack and recommendations for practical application.
AFRIKAANSE OPSOMMING: Gedurende die afgelope paar jaar was daar ‘n toename in oudio-ingenieurskursusse, insluitend studiemodules in oudio post-produksie, en selfs ‘n aanbod vir modules in post-produksie as hoofstudierigting. Desnieteenstaande, en ten spyte van die groei in die plaaslike filmindustrie is tot onlangs min akademiese studies op dié terrein in Suid-Afrika onderneem. In 2005 is ‘n MMus-tesis aan die Universiteit van KwaZulu-Natal voorgelê, met die titel Acoustic Ambience in Cinematography: An Exploration of the Descriptive and Emotional Impact of the Aural Environment (Turner, 2005: aanlyn). Hierdie tesis gee ‘n basiese oorsig oor die basiese komponente van die klankbaan, en fokus op die beskrywing en analise van die eienskappe van ambience – ‘n onderafdeling van klankeffekte. By die Universiteit van Stellenbosch is onlangs ‘n begin gemaak met navorsing oor die terreine van filmmusiek en Foley, d.w.s. klankeffekte geassosieer met menslike bewegings op die skerm.. Hierdie tesis beoog om ‘n oorsig te gee van oudio post-produksie en die bydrae van klank tot die filmmedium. Dit verskaf ‘n oorsig oor die prosesse betrokke by die daarstelling van ‘n filmklankbaan en sluit ook in ‘n beskrywing van die komponente van die klankbaan en aanbevelings vir die praktiese toepassing daarvan.
38
Lateef, Fehraz Aaron. "High fashion." Thesis, Queensland University of Technology, 2010. https://eprints.qut.edu.au/31818/1/Fehraz_Lateef_Thesis.pdf.
Full textAbstract:
High Fashion is a practice-led research enquiry that examines the processes involved in producing a no-budget film of high aesthetic standards that can confidently compete in the global film festival market, and to reflect on the production techniques tested during the making of the film.
The practical outcome of the research is a twenty-five minute short drama. It incorporates a large cast and crew, original designer clothing, extravagant sets, and a popular soundtrack. The thesis considers how over one hundred professionals volunteered their time, expertise, and equipment to help produce the film. The thesis also examines the many obstacles encountered while producing the film and how these were overcome. It is written for the student filmmaker as a guide to "learn by doing."
39
Yilmaz, Alper. "Translational control of mRNAs transcribed from HIV-1 provirus and HIV-1 based lentiviral vectors." Columbus, Ohio : Ohio State University, 2007. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1189785802.
Full text
40
Valotkaite, Justina. "Error detection for post-editing rule-based machine translation." Master's thesis, 2012. http://hdl.handle.net/10400.1/11064.
Full textAbstract:
The increasing role of Post-editing (PE) as a way of improving Machine Translation (MT) output and a faster alternative to translating from scratch among translators has lately attracted researchers’ attention. A number of recent studies have proposed various attempts to facilitate this task, especially for the outputs of Statistical
Machine Translation (SMT). However, little attention in the field has been given to Rule-based Machine Translation (RBMT).
In this dissertation an effort was made to provide support for the PE task through Error Detection (ED). A deep linguistic error analysis was done in a sample of English sentences in two text domains translated from Portuguese by two RBMT systems. The hypothesis is that automatically identifying and highlighting errors in
translations can help to perform the PE task faster, make it more efficient and less tedious.
As RBMT systems tend to make repetitive, systematic mistakes translators are forced to post-edit the same mistakes which makes their task monotonous and frustrating. In order to solve this problem, a set of 40 contrastive rules was designed tackling various linguistic phenomena on the basis of the translation errors identified in the error analysis. By applying this linguistic approach the project aimed at demonstrating that one can have a rule-based system working on the basis of designed rules which could help to detect and highlight translation errors in the RBMT output. The rules were verified by performing an experimental error analysis on a new data set whose results revealed that their coverage was 98.21%. The implementation results demonstrated a successful performance of the system. In addition, the results of a psycholinguistic experiment performed with human translators confirmed that having highlighted errors is useful as this can help translators perform the postediting task up to 12 seconds per error faster and improve their efficiency by minimizing the number of missed errors.
41
Rosa, Rudolf. "Automatic post-editing of phrase-based machine translation outputs." Master's thesis, 2013. http://www.nusl.cz/ntk/nusl-330262.
Full textAbstract:
We present Depfix, a system for automatic post-editing of phrase-based English-to-Czech machine trans- lation outputs, based on linguistic knowledge. First, we analyzed the types of errors that a typical machine translation system makes. Then, we created a set of rules and a statistical component that correct errors that are common or serious and can have a potential to be corrected by our approach. We use a range of natural language processing tools to provide us with analyses of the input sentences. Moreover, we reimple- mented the dependency parser and adapted it in several ways to parsing of statistical machine translation outputs. We performed both automatic and manual evaluations which confirmed that our system improves the quality of the translations.
42
Schumann, Paula. "Post-Editing als Bestandteil von Translationsstudiengängen in der DACH-Region." 2020. https://ul.qucosa.de/id/qucosa%3A39018.
Full text
43
Scholz, Volker [Verfasser]. "New editing techniques for video post-processing / vorgelegt von Volker Scholz." 2007. http://d-nb.info/985574437/34.
Full text
44
Fernandes, Pedro Miguel Sousa. "Community-based Sports Articles Generation Platform using NLG and Post-Editing." Master's thesis, 2021. https://hdl.handle.net/10216/135617.
Full text
45
Fernandes, Pedro Miguel Sousa. "Community-based Sports Articles Generation Platform using NLG and Post-Editing." Dissertação, 2021. https://hdl.handle.net/10216/135617.
Full text
46
Comparin, Lucia. "Quality in machine translation and human post-editing : error annotation and specifications." Master's thesis, 2016. http://hdl.handle.net/10451/27969.
Full textAbstract:
Machine translation (MT) has been an important field of research in the last decades and is currently playing a key role in the translation market. The variable quality of results depending on various factors makes it necessary to combine MT with post-editing, to obtain high-quality translation. Post-editing is, nonetheless, a costly and time-consuming task. In order to improve the overall performance of a translation workflow involving MT, it is crucial to evaluate the quality of results produced to identify the main errors and outline strategies to address them. In this study, we assessed the results of MT and after the first human post-edition at Unbabel, a Portuguese startup that provides translation services combining MT with post-editing performed online by a community of editors. A corpus of texts translated at Unbabel from English into Italian was annotated after MT and after the first post-edition step. The data collected allowed us to identify three types of errors that are frequent and critical in terms of quality, namely “word order”, “agreement”, and “tense/mood/aspect”. Hence, correcting the errors belonging to these categories would have a major impact on the quality of translation and turn the post-editing process more accurate and efficient. The errors annotated in the corpus were analyzed in order to identify common patterns of errors, and possible solutions to address the issues identified were outlined. The MT system used at Unbabel and the tools available determined the choice to integrate information retrieved by error analysis in the Smartcheck, the tool used at Unbabel to automatically detect errors in the target text produced by the MT system and provide relevant messages to the editors. Therefore, our study focused on the definition and integration of rules in the Smartcheck to detect the most frequent and critical errors in the texts, in order to provide informative and accurate messages to the editor to aid him/her in the post-editing process.A tradução automática tem vindo a assumir uma grande importância no mercado da tradução e representa atualmente uma importante área de investigação. Durante os últimos cinquenta anos, vários sistemas de tradução automática foram desenvolvidos com base em paradigmas e abordagens diferentes. Os sistemas de tradução automática podem ser divididos entre sistemas baseados em conhecimento linguístico em forma de regras e sistemas baseados em corpora de textos, como os estatísticos e os baseados em exemplos. Além disso, nas últimas décadas, paradigmas diferentes foram combinados para desenvolver sistemas híbridos que utilizam corpora de textos, como nos sistemas estatísticos ou nos baseados em exemplos, mas integram regras e princípios linguísticos, como nos sistemas baseados em conhecimento, para resolver dificuldades gramaticais ou lexicais. Os sistemas de tradução automática são cada vez mais utilizados no processo de tradução, devido ao crescente volume de textos para traduzir e aos curtos prazos estabelecidos. Apesar de haver diferentes sistemas, os resultados são variáveis no que diz respeito à qualidade, dependendo do paradigma e do grau de especialização do sistema e dos textos a traduzir num determinado domínio. Estes factos impõem a necessidade de realizar uma edição dos textos, que pode ocorrer antes da tradução (pré-edição) ou depois (pós-edição). No primeiro caso, do texto de partida são eliminadas as estruturas ou palavras que representam dificuldades para a tradução automática realizada por um sistema em particular. No segundo caso, o texto traduzido pelo sistema é controlado e corrigido por um revisor humano. Para que este tipo de processo possa ser utilizado em grande escala no mercado da tradução, é importante reduzir os custos que lhe são inerentes e agilizá-lo. Além da pré-edição ou pós-edição, em função do paradigma considerado, integrar mais informação linguística ou atualizar os recursos lexicais utilizados permite melhorar os resultados da tradução automática. O presente trabalho tem como objeto de estudo o controlo de qualidade na área da tradução automática, mais especificamente, na fase de pós-edição. O estudo e a análise dos resultados da tradução automática e da fase de pós-edição permitem delinear estratégias para intervir em dois sentidos: por um lado, melhorar os resultados do sistema de tradução automática graças à integração de mais informação no sistema; por outro lado, apoiar o trabalho do revisor na pós-edição, destacando erros prováveis ou assinalando pontos críticos. A avaliação dos resultados da tradução automática inclui uma fase de análise dos erros presentes no texto de chegada e uma classificação dos mesmos, de acordo com uma tipologia de categorias de erros. No estudo da fase de pós-edição, a análise dos erros mais frequentemente corrigidos pelos revisores permite identificar que tipo de informação deve ser integrada no sistema de tradução automática e que instruções podem ser úteis aos revisores. Para a realização desta análise, adotou-se um sistema de classificação a fim de categorizar os erros e, portanto, de realizar uma avaliação quantitativa da qualidade da tradução. O presente trabalho de projeto foi realizado em colaboração com a Unbabel, uma startup portuguesa que oferece serviços de tradução quase em tempo real, combinando tradução automática com uma comunidade de revisores. O corpus que é utilizado para a realização do trabalho que aqui se propõe é formado por textos em língua inglesa, traduzidos para italiano através de um sistema de tradução automática, corrigidos e editados por vários revisores humanos. São analisados os erros presentes nos textos de chegada após a tradução automática e a primeira revisão. A identificação e a análise dos erros permite chegar a generalizações sob a forma de regras a ser implementadas no processo tradução e pós-edição de textos realizado pela Unbabel. Em particular, as regras destinam-se à integração numa ferramenta que identifica automaticamente os erros no texto de chegada de algumas categorias específicas, depois da tradução automática e durante o processo de pós-edição. A ferramenta assinala o erro e, em função do tipo de problema, sugere ao revisor uma correção ou dá-lhe indicações para prestar atenção a um aspeto particular da sequência assinalada, pois é provável que contenha um erro. O presente trabalho divide-se em oito capítulos em que são abordados os temas fundamentais envolvidos na realização do trabalho. No primeiro capítulo apresenta-se o objeto de estudo, a motivação do trabalho de projeto, a abordagem metodológica adoptada e a organização do documento. No segundo capítulo apresenta-se a fundamentação teórica em que se baseou o estudo. Aborda-se brevemente a história da tradução automática, desde as suas primeiras tentativas em meados do século XX, até aos mais recentes sistemas da primeira década do século XXI. Após a apresentação da história, são descritas algumas dificuldades linguísticas e operacionais relacionadas com a tradução automática e apresenta-se uma descrição dos diferentes sistemas de tradução automática, nomeadamente os baseados em conhecimento linguístico, os baseados em corpora e os híbridos. No terceiro capítulo apresenta-se o processo de tradução automática utilizado na Unbabel, fazendo-se uma breve descrição dos passos que o compõem, o sistema de tradução automática usado para a tradução dos textos do corpus e as ferramentas utilizadas na fase de pós-edição para a deteção de erros e para os testes de qualidade. No quarto capítulo introduz-se a tarefa da anotação de erros descrevendo-se, em primeiro lugar, a tipologia de erros adotada na análise e a ferramenta usada para a tarefa. Seguidamente, é apresentado o corpus de textos considerado neste estudo e são apresentados os dados recolhidos, nomeadamente o número de erros anotados nos textos depois da tradução automática e depois da primeira fase de pós-edição. Uma análise do número de erros anotados nas várias categorias de erros segue-se a apresentação dos dados e justifica a escolha de algumas categorias de erros para as quais são propostas soluções. Nos três capítulos seguintes são analizados os erros que pertencem às três categorias escolhidas, nomeadamente “word order” (ordem de palavras), no quinto capítulo, “agreement” (concordância), no sexto capítulo, e “tense/mood/aspect” (tempo/modo/aspeto), no sétimo capítulo. Em primeiro lugar, para cada categoria de erro, são abordadas as linhas gerais que caraterizam o fenómeno linguístico em inglês e italiano, e em seguida, os erros anotados são analisados e divididos em sub-categorias. Isto permite encontrar padrões de erros frequentes e generalizá-los, de maneira a poder propor soluções gerais que dêem conta de todos os erros do mesmo tipo. No último capítulo apresentam-se as conclusões e o trabalho futuro que pode ser realizado como continuação do presente estudo e aproveitando aspetos que não foi possível explorar no âmbito do trabalho de projeto aqui apresentado. Em suma, o presente trabalho centra-se na identificação de questões problemáticas e na proposta de soluções para a melhoria da qualidade dos resultados no processo de tradução automática, na fase de pós-edição, constituindo um importante contributo não só para a formação da mestranda no âmbito dos sistemas de tradução automática e do seu funcionamento, como também para a melhoria do desempenho do sistema de trabalho específico levado a cabo na Unbabel.
47
Sousa, Sheila Castilho Monteiro de. "Automatic and semi-automatic translation of DVD subtitle: measuring post-editing effort." Master's thesis, 2011. http://hdl.handle.net/10400.1/10633.
Full textAbstract:
Dissertação de Mestrado, Processamento de Linguagem Natural e Tecnologia da Linguagem Humana, Faculdade de Ciências Humanas e Sociais, Universidade do Algarve, 2011A crescente demanda de tradução audiovisual e o curto tempo que os legendadores tem para realizar a tradução fizeram da Tradução Automática e da Memória de Tradução potenciais ajudas para os profissionais nesse campo. Em recentes publicações tentaram-se traduzir legendas automaticamente porém nenhuma delas tentou-se medir o esforço para pós-editar a saída dos sistemas. Medir o esforço é um passo essencial para avaliar se o sistema de tradução automática pode realmente ajudar o trabalho dos legendadores. Neste trabalho, mede-se o esforço para pós-editar as saídas de um sistema baseado em regras, dois sistemas estatísticos e também de uma memória de tradução ao traduzir um corpus de Inglês para o Português Brasileiro. Onze tradutores avaliaram as sentenças de acordo com o esforço para traduzir e pós-editar fazendo o uso de uma ferramenta que computa o tempo gasto em cada tarefa. Os resultados mostram que o tempo médio necessário para traduzir uma frase do zero é, em média, 70% mais lento do que pós-editar a mesma frase. De acordo com avaliação feita pelos tradutores, mais de 69% das sentenças de saída necessitaram pouca ou nenhuma pós-edição.
48
Carmo, Félix Emanuel Martins do. "post-editing: a theoretical and practical challenge for translation studies and machine learning." Doctoral thesis, 2017. https://repositorio-aberto.up.pt/handle/10216/107518.
Full text
49
Carmo, Félix Emanuel Martins do. "post-editing: a theoretical and practical challenge for translation studies and machine learning." Tese, 2017. https://repositorio-aberto.up.pt/handle/10216/107518.
Full text
50
Hung, Chao-Yi, and 洪兆怡. "A Study of the Keys to Better Post-Editing in Machine-Aided Human Translation." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/29328539421168030795.
Full textAbstract:
碩士國立臺灣科技大學
應用外語系
104
Machine translation (MT) has been gone along for over sixty years, and gradually develops a translation combination of Machine Translation (MT) and Post-Editing (PE). Although this combination successfully increases the productivity and quality of translation, some problems, like word sense disambiguation and reordering, derived from MT still make PE efficiency impossible. In order to solve above MT problems, this study raises up four study questions: (1) what are the major problems that affect the quality of current MT? (2) what are the bottlenecks that compromise the efficiency of post-editing in MAHT? (3) what kinds of training are needed to enable post-editors to address the problems derived from MT’s poor reordering? (4) how can a language parser help with the training to untangle cluttered output texts from MT in post-editing training? This study uses Google Translate (GT) as the testing model to firstly produce the output out of different sample texts. Then the GT texts were compared with the original texts (OT) parsed with a language parser to identify the errors between GT and OT. Finally, those problems were summarized and, as a result, produce a checklist of 40 checkpoints and 7 guidelines. Moreover, around 19 training goals are laid out for the reference of post-editor training programs. According to the results, four research questions of this study are answered: (1) the major problems that affect the quality of current MT are WSD and reordering, mainly reordering; (2) the bottlenecks that compromise the efficiency of post-editing in MAHT is reordering; (3) instilling and enhancing post-editors’ perspective of linguistic are needed to address the problems derived from MT’s poor reordering; (4) the guideline and checkpoints can help the training to untangle cluttered output texts from MT in post-editing training. All in all, with the aid of the checklist, the translators will gradually improve their PE skills, as well as the productivity and quality of translation.