Search

[EN] The problem of automatic text classification is an essential part of text analysis. The improvement of text classification can be done at different levels such as a preprocessing step, network implementation, etc. In this paper, we focus on how the combination of different methods of text encoding may affect classification accuracy. To do this, we implemented a multi-input neural network that is able to encode input text using several text encoding techniques such as BERT, neural embedding layer, GloVe, skip-thoughts and ParagraphVector. The text can be represented at different levels of tokenised input text such as the sentence level, word level, byte pair encoding level and character level. Experiments were conducted on seven datasets from different language families: English, German, Swedish and Czech. Some of those languages contain agglutinations and grammatical cases. Two out of seven datasets originated from real commercial scenarios: (1) classifying ingredients into their corresponding classes by means of a corpus provided by Northfork; and (2) classifying texts according to the English level of their corresponding writers by means of a corpus provided by ProvenWord. The developed architecture achieves an improvement with different combinations of text encoding techniques depending on the different characteristics of the datasets. Once the best combination of embeddings at different levels was determined, different architectures of multi-input neural networks were compared. The results obtained with the best embedding combination and best neural network architecture were compared with state-of-the-art approaches. The results obtained with the dataset used in the experiments were better than the state-of-the-art baselines.

[EN] One of the most popular approaches to machine translation consists in formulating the problem as a pattern recognition approach. Under this perspective, bilingual corpora are precious resources, as they allow for a proper estimation of the underlying models. In this framework, selecting the best possible corpus is critical, and data selection aims to find the best subset of the bilingual sentences from an available pool of sentences such that the final translation quality is improved. In this paper, we present a new data selection technique that leverages a continuous vector-space representation of sentences. Experimental results report improvements compared not only with a system trained only with in-domain data, but also compared with a system trained on all the available data. Finally, we compared our proposal with other state-of-the-art data selection techniques (Cross-entropy selection and Infrequent ngrams recovery) in two different scenarios, obtaining very promising results with our proposal: our data selection strategy is able to yield results that are at least as good as the best-performing strfategy for each scenario. The empirical results reported are coherent across different language pairs.

[EN] We present a simple and reliable method for estimating the log-linear weights of a state-of-the-art machine translation system, which takes advantage of the method known as discriminative ridge regression (DRR). Since inappropriate weight estimations lead to a wide variability of translation quality results, reaching a reliable estimate for such weights is critical for machine translation research. For this reason, a variety of methods have been proposed to reach reasonable estimates. In this paper, we present an algorithmic description and empirical results proving that DRR is able to provide comparable translation quality when compared to state-of-the-art estimation methods [i.e. MERT and MIRA], with a reduction in computational cost. Moreover, the empirical results reported are coherent across different corpora and language pairs.

[ES] La Traducción Automática Estadística es un sup-campo de la lingüística computacional que investiga como emplear los ordenadores en el proceso de traducción de un texto de un lenguaje humano a otro. La traducción automática estadística es el enfoque más popular que se emplea para construir estos sistemas de traducción automáticos. La calidad de dichos sistemas depende en gran medida de los ejemplos de traducción que se emplean durante los procesos de entrenamiento y adaptación de los modelos. Los conjuntos de datos empleados son obtenidos a partir de una gran variedad de fuentes y en muchos casos puede que no tengamos a mano los datos más adecuados para un dominio específico. Dado este problema de carencia de datos, la idea principal para solucionarlo es encontrar aquellos conjuntos de datos más adecuados para entrenar o adaptar un sistema de traducción. En este sentido, esta tesis propone un conjunto de técnicas de selección de datos que identifican los datos bilingües más relevantes para una tarea extraídos de un gran conjunto de datos. Como primer paso en esta tesis, las técnicas de selección de datos son aplicadas para mejorar la calidad de la traducción de los sistemas de traducción bajo el paradigma basado en frases. Estas técnicas se basan en el concepto de representación continua de las palabras o las oraciones en un espacio vectorial. Los resultados experimentales demuestran que las técnicas utilizadas son efectivas para diferentes lenguajes y dominios. El paradigma de Traducción Automática Neuronal también fue aplicado en esta tesis. Dentro de este paradigma, investigamos la aplicación que pueden tener las técnicas de selección de datos anteriormente validadas en el paradigma basado en frases. El trabajo realizado se centró en la utilización de dos tareas diferentes de adaptación del sistema. Por un lado, investigamos cómo aumentar la calidad de traducción del sistema, aumentando el tamaño del conjunto de entrenamiento. Por otro lado, el método de sele, [CA] La Traducció Automàtica Estadística és un sup-camp de la lingüística computacional que investiga com emprar els ordinadors en el procés de traducció d'un text d'un llenguatge humà a un altre. La traducció automàtica estadística és l'enfocament més popular que s'empra per a construir aquests sistemes de traducció automàtics. La qualitat d'aquests sistemes depèn en gran mesura dels exemples de traducció que s'empren durant els processos d'entrenament i adaptació dels models. Els conjunts de dades emprades són obtinguts a partir d'una gran varietat de fonts i en molts casos pot ser que no tinguem a mà les dades més adequades per a un domini específic. Donat aquest problema de manca de dades, la idea principal per a solucionar-ho és trobar aquells conjunts de dades més adequades per a entrenar o adaptar un sistema de traducció. En aquest sentit, aquesta tesi proposa un conjunt de tècniques de selecció de dades que identifiquen les dades bilingües més rellevants per a una tasca extrets d'un gran conjunt de dades. Com a primer pas en aquesta tesi, les tècniques de selecció de dades són aplicades per a millorar la qualitat de la traducció dels sistemes de traducció sota el paradigma basat en frases. Aquestes tècniques es basen en el concepte de representació contínua de les paraules o les oracions en un espai vectorial. Els resultats experimentals demostren que les tècniques utilitzades són efectives per a diferents llenguatges i dominis. El paradigma de Traducció Automàtica Neuronal també va ser aplicat en aquesta tesi. Dins d'aquest paradigma, investiguem l'aplicació que poden tenir les tècniques de selecció de dades anteriorment validades en el paradigma basat en frases. El treball realitzat es va centrar en la utilització de dues tasques diferents. D'una banda, investiguem com augmentar la qualitat de traducció del sistema, augmentant la grandària del conjunt d'entrenament. D'altra banda, el mètode de selecció de dades es va emprar per a crear un conjunt de dades, [EN] La Traducció Automàtica Estadística és un sup-camp de la lingüística computacional que investiga com emprar els ordinadors en el procés de traducció d'un text d'un llenguatge humà a un altre. La traducció automàtica estadística és l'enfocament més popular que s'empra per a construir aquests sistemes de traducció automàtics. La qualitat d'aquests sistemes depèn en gran mesura dels exemples de traducció que s'empren durant els processos d'entrenament i adaptació dels models. Els conjunts de dades emprades són obtinguts a partir d'una gran varietat de fonts i en molts casos pot ser que no tinguem a mà les dades més adequades per a un domini específic. Donat aquest problema de manca de dades, la idea principal per a solucionar-ho és trobar aquells conjunts de dades més adequades per a entrenar o adaptar un sistema de traducció. En aquest sentit, aquesta tesi proposa un conjunt de tècniques de selecció de dades que identifiquen les dades bilingües més rellevants per a una tasca extrets d'un gran conjunt de dades. Com a primer pas en aquesta tesi, les tècniques de selecció de dades són aplicades per a millorar la qualitat de la traducció dels sistemes de traducció sota el paradigma basat en frases. Aquestes tècniques es basen en el concepte de representació contínua de les paraules o les oracions en un espai vectorial. Els resultats experimentals demostren que les tècniques utilitzades són efectives per a diferents llenguatges i dominis. El paradigma de Traducció Automàtica Neuronal també va ser aplicat en aquesta tesi. Dins d'aquest paradigma, investiguem l'aplicació que poden tenir les tècniques de selecció de dades anteriorment validades en el paradigma basat en frases. El treball realitzat es va centrar en la utilització de dues tasques diferents d'adaptació del sistema. D'una banda, investiguem com augmentar la qualitat de traducció del sistema, augmentant la grandària del conjunt d'entrenament. D'altra banda, el mètode de selecció de dades es va emprar per a cr

[EN] The filtering task of noisy parallel corpora in WMT2019 aims to challenge participants to create filtering methods to be useful for training machine translation systems. In this work, we introduce a noisy parallel corpora filtering system based on generating hypotheses by means of a translation model. We train translation models in both language pairs: Nepali English and Sinhala English using provided parallel corpora. To create the best possible translation model, we first join all provided parallel corpora (Nepali, Sinhala and Hindi to English) and after that, we applied bilingual cross-entropy selection for both language pairs (Nepali English and Sinhala English). Once the translation models are trained, we translate the noisy corpora and generate a hypothesis for each sentence pair. We compute the smoothed BLEU score between the target sentence and generated hypothesis. In addition, we apply several rules to discard very noisy or inadequate sentences which can lower the translation score. These heuristics are based on sentence length, source and target similarity and source language detection. We compare our results with the baseline published on the shared task website, which uses the Zipporah model, over which we achieve significant improvements in one of the conditions in the shared task. The designed filtering system is domain independent and all experiments are conducted using neural machine translation.

Neural Machine Translation (NMT) has achieved promising results comparable with Phrase-Based Statistical Machine Translation (PBSMT). However, to train a neural translation engine, much more powerful machines are required than those required to develop translation engines based on PBSMT. One solution to reduce the training cost of NMT systems is the reduction of the training corpus through data selection (DS) techniques. There are many DS techniques applied in PBSMT which bring good results. In this work, we show that the data selection technique based on infrequent n-gram occurrence described in (Gascó et al., 2012) commonly used for PBSMT systems also works well for NMT systems. We focus our work on selecting data according to specific corpora using the previously mentioned technique. The specific-domain corpora used for our experiments are IT domain and medical domain. The DS technique significantly reduces the execution time required to train the model between 87% and 93%. Also, it improves translation quality by up to 2.8 BLEU points. The improvements are obtained with just a small fraction of the data that accounts for between 6% and 20% of the total data.

[EN] We present a simple and reliable method for estimating the log-linear weights of a state-of-the-art machine translation system, which takes advantage of the method known as discriminative ridge regression (DRR). Since inappropriate weight estimations lead to a wide variability of translation quality results, reaching a reliable estimate for such weights is critical for machine translation research. For this reason, a variety of methods have been proposed to reach reasonable estimates. In this paper, we present an algorithmic description and empirical results proving that DRR, as applied in a pseudo-batch scenario, is able to provide comparable translation quality when compared to state-of-the-art estimation methods (i.e., MERT [1] and MIRA [2]). Moreover, the empirical results reported are coherent across different corpora and language pairs.

[EN] Domain adaptation has recently gained interest in statistical machine translation. One of the adaptation techniques is based in the selection data. Data selection aims to select the best subset of the bilingual sentences from an available pool of sentences, with which to train a SMT system. In this paper, we study how affect the bilingual corpora used for the data selection methods in the translation quality, [ES] La adaptación de dominios genera mucho interés dentro de la traducción automática estadística. Una de las técnicas de adaptaciión esta basada en la selecciión de datos que tiene como objetivo seleccionar el mejor subconjunto de oraciones bilingües de un gran conjunto de oraciones. En este artículo estudiamos como afectan los corpus bilingües empleados por los métodos de selección de frases en la calidad de las traducciones.

Globalization suddenly brings many people from different country to interact with each other, requiring them to be able to speak several languages. Human translators are slow and expensive, we find the necessity of developing machine translators to automatize the task. Several approaches of Machine translation have been develop by the researchers. In this work, we use the Statistical Machine Translation approach. Statistical Machine Translation systems perform poorly when applied on new domains. The domain adaptation problem has recently gained interest in Statistical Machine Translation. The basic idea is to improve the performance of the system trained and tuned with different domain than the one to be translated. This article studies different paradigms of domain adaptation. The results report improvements compared with a system trained only with in-domain data and trained with all the available data.

[EN] We introduce a Bayesian approach for the adaptation of the log-linear weights present in state-of-the-art statistical machine translation systems. Typically, these weights are estimated by optimising a given translation quality criterion, taking only into account a certain set of development data (e.g., the adaptation data). In this article, we show that the Bayesian framework provides appropriate estimates of such weights in conditions where adaptation data is scarce. The theoretical framework is presented, alongside with a thorough experimentation and comparison with other weight estimation methods. We provide a comparison of different sampling strategies, including an effective heuristic strategy and a theoretically sound Markov chain Monte-Carlo algorithm. Experimental results show that Bayesian predictive adaptation (BPA) outperforms the re-estimation from scratch in conditions where adaptation data is scarce. Further analysis reveals that the improvements obtained are due to the greater stability of the estimation procedure. In addition, the proposed BPA framework has a much lower computational cost than raw re-estimation. © 2015 Elsevier Ltd. All rights reserved.

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-19390-8_49, In this paper, we present a clustering approach based on the combined use of a continuous vector space representation of sentences and the k-means algorithm. The principal motivation of this proposal is to split a big heterogeneous corpus into clusters of similar sentences. We use the word2vec toolkit for obtaining the representation of a given word as a continuous vector space. We provide empirical evidence for proving that the use of our technique can lead to better clusters, in terms of intra-cluster perplexity and F 1 score.

[EN] In text entry experiments, memorability is a desired property of the phrases used as stimuli. Unfortunately, to date there is no automated method to achieve this effect. As a result, researchers have to use either manually curated Englishonly phrase sets or sampling procedures that do not guarantee phrases being memorable. In response to this need, we present a novel sampling method based on two core ideas: a multiple regression model over language-independent features, and the statistical analysis of the corpus from which phrases will be drawn. Our results show that researchers can finally use a method to successfully curate their own stimuli targeting potentially any language or domain. The source code as well as our phrase sets are publicly available.

[ES] La intervención de los traductores humanos en un escenario de post-edición para corregir las traducciones obtenidas a partir de los sistemas de traducción automática es aún muy necesaria para lograr la calidad deseada. El paradigma de la traducción automática interactiva (Interactive Machine Translation, IMT), es capaz de reducir el esfuerzo y tiempo que el traductor humano tiene que invertir en el proceso de corrección. En este trabajo final de máster se plantea la utilización del paradigma de traducción automática interactiva, combinado con una aproximación que adecua los pesos del modelo log-lineal a cada una de las traducciones mediante diferentes algoritmos de aprendizaje online. Nuestro objetivo es que el sistema aprenda de los errores corregidos, favoreciendo la corrección de las próximas traducciones. Para lograr lo anteriormente planteado se emplearon diferentes algoritmos de aprendizaje online: Discriminative Ridge Regression, Perceptron-Like y Passive Agressive, empleados estos en postedición con resultados positivos. Para poder utilizar estos algoritmos dentro del escenario IMT fue necesaria una nueva formulación de cada uno de los algoritmos. Con estas nuevas formulaciones, en este trabajo final de máster, se obtienen resultados diversos, dando la posibilidad de emplearse en nuevos planteamientos para lograr la calidad de las traducciones deseada y así disminuir el esfuerzo del traductor humano., [EN] In a post-edit scenario, the translations obtained by machine translator systems need to have been corrected by a human translator to obtain the desire quality. Interactive Machine Tranlator (IMT) paradigm is able to reduce the effort and the time that human translators have to invert in the correction process. In this thesis, we propose to adapt the weights of the log-linear model in interactive machine translator. For adapting the weights of the log-linear model, we have utilizes different online learning algorithms. The main goal is that the system learns from the errors corrected. We propose to use three different online learning algorithms: Discriminative Ridge Regression, Passive Agressive and Percetron-Like. These algorithms has been used in post-edit scenario with good results. These algorithms needed a new formulation in IMT sceneario. With these new formulations, we have obtained different results. These resuts give the posibility to use the new formulations to archieve the quality deseared and reduce efforts of the human translator in new problems.