Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 19
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Active learning for interactive machine translation

  • González Rubio, Jesús
  • Ortiz Martínez, Daniel
  • Casacuberta Nolla, Francisco|||0000-0002-8497-5598
Translation needs have greatly increased during the last years. In many situations, text to be translated constitutes an unbounded stream of data that grows continually with time. An effective approach to translate text documents is to follow an interactive-predictive paradigm in which both the system is guided by the user and the user is assisted by the system to generate error-free translations. Unfortunately, when processing such unbounded data streams even this approach requires an overwhelming amount of manpower. Is in this scenario where the use of active learning techniques is compelling. In this work, we propose different active learning techniques for interactive machine translation. Results show that for a given translation quality the use of active learning allows us to greatly reduce the human effort required to translate the sentences in the stream.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Passive-aggressive for on-line learning in statistical machine translation

  • Martínez Gómez, Pascual
  • Sanchis Trilles, Germán
  • Casacuberta Nolla, Francisco|||0000-0002-8497-5598
New variations on the application of the passive-aggressive algorithm to statistical machine translation are developed and compared to previously existing approaches. In online adaptation, the system needs to adapt to real-world changing scenarios, where training and tuning only take place when the system is set-up for the first time. Post-edit information, as described by a given quality measure, is used as valuable feedback within the passive-aggressive framework, adapting the statistical models on-line. First, by modifying the translation model parameters, and alternatively, by adapting the scaling factors present in stateof- the-art SMT systems. Experimental results show improvements in translation quality by allowing the system to learn on a sentence-by-sentence basis., This paper is based upon work supported by the EC (FEDER/FSE) and the Spanish MICINN under projects MIPRCV “Consolider Ingenio 2010” (CSD2007-00018) and iTrans2 (TIN2009-14511). Also supported by the Spanish MITyC under the erudito.com (TSI-020110-2009-439) project, by the Generalitat Valenciana under grant Prometeo/2009/014 and scholarship GV/2010/067 and by the UPV under grant 20091027.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Multimodal interactive transcription of ancient text images

  • Romero Gómez, Verónica
  • Sánchez Peiró, Joan Andreu|||0000-0003-0423-2020
  • Toselli, Alejandro Héctor|||0000-0001-6955-9249
  • Vidal, Enrique|||0000-0003-4579-5196
The amount of digitized legacy documents has been rising dramatically over the last years due mainly to the increasing number of on-line digital libraries publishing this kind of documents. On one hand, the vast majority of these documents remain waiting to be transcribed into a textual electronic format (such as ASCII or PDF) that would provide historians and other researchers new ways of indexing, consulting and querying these documents. On the other hand, in some cases, adequate transcriptions of the handwritten text images are already available. This drives an increasing need to align images and transcriptions in order to make it more comfortable the consulting of these documents. In this work two systems are presented to deal with these issues. The first one aims at transcribing these documents using a interactive-predictive approach, which integrates user corrective-feedback actions in the proper recognition process. The second one presents an alignment method based on the Viterbi algorithm to find mappings between word images of a given handwritten document and their respective (ASCII) words on its given transcription., Work supported by the Spanish Government (MICINN and “Plan E”) under the MITTRAL (TIN2009-14633-C03-01) research project and under the research programme Consolider Ingenio 2010: MIPRCV (CSD2007- 00018) and the Generalitat Valenciana under gran Prometeo/2009/14.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Active learning for dialogue act labelling

  • Ghigi, Fabrizio
  • Tamarit Ballester, Vicent
  • Benedí Ruiz, José Miguel
  • Martínez-Hinarejos, Carlos-D.|||0000-0002-6139-2891
Active learning is a useful technique that allows for a considerably reduction of the amount of data we need to manually label in order to reach a good performance of a statistical model. In order to apply active learning to a particular task we need to previously define an effective selection criteria, that picks out the most informative samples at each iteration of active learning process. This is still an open problem that we are going to face in this work, in the task of dialogue annotation at dialogue act level. We present two different criteria, weighted number of hypothesis and entropy, that we have applied to the Sample Selection Algorithm for the task of dialogue act labelling, that retrieved appreciably improvements in our experimental approach. © 2011 Springer-Verlag., Work supported by the EC (FEDER/FSE) and the Spanish MEC/MICINN under the MIPRCV “Consolider Ingenio 2010” program (CSD2007-00018), MITTRAL (TIN2009-14633-C03-01) projects and the FPI scholarship (BES-2009-028965). Also supported by the Generalitat Valenciana under grant Prometeo/2009/014 and GV/2010/067
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Language technology for handwritten text recognition

  • Toselli, Alejandro Héctor|||0000-0001-6955-9249
  • Giménez Pastor, Adrián
  • Juan, Alfons|||0000-0002-9984-4072
  • Vidal, Enrique|||0000-0003-4579-5196
  • Serrano Martínez-Santos, Nicolás
  • Khoury, Ihab
This paper shows how the nowadays prevalent technology used in HTR borrows concepts and methods from the field of ASR; i.e. those based on Hidden Markov Models (HMMs). Additionally, it will be described a HTR approach based on employing Bernoulli distributions rather than Gaussian-Mixture distributions for the HMM-state emission probability of observations. Finally, handwritten text recognition evaluation results are reported for several corpora involving different characteristics and languages., Work supported by the EC (FEDER), the Spanish MEC under the MIPRCV “Consolider Ingenio 2010” research programme (CSD2007- 00018) and the Spanisg Government (MICINN and “Plan E”) under the MITTRAL (TIN2009-14633-C03-01) research project.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Structural graph extraction from images

  • Gallego-Sánchez, Antonio-Javier
  • Calera Rubio, Jorge
  • López Rodríguez, Damián|||0000-0003-3633-3862
We present three new algorithms to model images with graph primitives. Our main goal is to propose algorithms that could lead to a broader use of graphs, especially in pattern recognition tasks. The first method considers the q-tree representation and the neighbourhood of regions. We also propose a method which, given any region of a q-tree, finds its neighbour regions. The second algorithm reduces the image to a structural grid. This grid is postprocessed in order to obtain a directed acyclic graph. The last method takes into account the skeleton of an image to build the graph. It is a natural generalization of similar works on trees [8, 12]. Experiments show encouraging results and prove the usefulness of the proposed models in more advanced tasks, such as syntactic pattern recognition tasks., This work is partially supported by Spanish MICINN (contract TIN2011-28260-C03-01, contract TIN2009-14205-C04-C1) and CONSOLIDER-INGENIO 2010 (contract CSD2007-00018)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Language identification for interactive handwriting transcription of multilingual documents

  • Del Agua Teba, Miguel Angel
  • Serrano Martinez Santos, Nicolas
  • Juan, Alfons|||0000-0002-9984-4072
An effective approach to handwriting transcription of (old) documents is to follow a sequential, line-by-line transcription of the whole document, in which a continuously retrained system interacts with the user. In the case of multilingual documents, however, a minor yet important issue for this interactive approach is to first identify the language of the current text line image to be transcribed. In this paper, we propose a probabilistic framework and three techniques for this purpose. Empirical results are reported on an entire 764-page multilingual document for which previous empirical tests were limited to its first 180 pages, written only in Spanish. © 2011 Springer-Verlag., Work supported by the EC (FEDER, FSE), the Spanish Government (MICINN, MITyC, “Plan E”, under grants MIPRCV “Consolider Ingenio 2010”, MITTRAL TIN2009- 14633-C03-01 and FPU AP2007-02867), the Generalitat Valenciana (grant Prometeo/2009/014 and ACOMP/2010/051) and the UPV (grant 20080033).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

On the use of diagonal and class-dependent weighted distances for the Probabilistic k-nearest neighbor

  • Paredes Palacios, Roberto
  • Girolami, Mark
A probabilistic k-nn (PKnn) method was introduced in [13] under the Bayesian point of view. This work showed that posterior inference over the parameter k can be performed in a relatively straightforward manner using Markov Chain Monte Carlo (MCMC) methods. This method was extended by Everson and Fieldsen [14] to deal with metric learning. In this work we propose two different dissimilarities functions to be used inside this PKnn framework. These dissimilarities functions can be seen as a simplified version of the full-covariance distance functions just proposed. Furthermore we propose to use a class- dependent dissimilarity function as proposed in [8] aim at improving the k-nn classifier. In the present work we pursue a simultaneously learning of the dissimilarity function parameters together with the parameter k of the k-nn classifier. The experiments show that this simultaneous learning lead to an improvement of the classifier with respect to the standard k-nn and state-of-the-art technique as well., Work supported by the Spanish MEC/MICINN under the MIPRCV Consolider Ingenio 2010 program (CSD2007-00018).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Explicit length modelling for statistical machine translation

  • Silvestre Cerdà, Joan Albert|||0000-0003-2291-8296
  • Civera Saiz, Jorge|||0000-0002-0963-0143
  • Andrés Ferrer, Jesús
Explicit length modelling has been previously explored in statistical pattern recognition with successful results. In this paper, two length models along with two parameter estimation methods for statistical machine translation (SMT) are presented. More precisely, we incorporate explicit length modelling in a state-of-the-art log-linear SMT system as an additional feature function in order to prove the contribution of length information. Finally, promising experimental results are reported on a reference SMT task. © 2011 Springer-Verlag., Work supported by the EC (FEDER/FSE) and the Spanish MEC/MICINN under the MIPRCV “Consolider Ingenio 2010” program (CSD2007-00018) and iTrans2 (TIN2009-14511) projects. Also supported by the Spanish MITyC under the erudito.com (TSI-020110-2009-439) project and by the Generalitat Valenciana under grant Prometeo/2009/014 and GV/2010/067, and by the “Vicerrectorado de Investigaci´on de la UPV” under grant 20091027.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Bernoulli HMMs at subword level for handwritten word recognition

  • Giménez Pastor, Adrián
  • Juan, Alfons
This paper presents a handwritten word recogniser based on HMMs at subword level (characters) in which state-emission probabilities are governed by multivariate Bernoulli probability functions. This recogniser works directly with raw binary pixels of the image, instead of conventional, real-valued local features. A detailed experimentation has been carried out by varying the number of states, and comparing the results with those from a conventional system based on continuous (Gaussian) densities. From this experimentation, it becomes clear that the proposed recogniser is much better than the conventional system, Work supported by the EC (FEDER) and the Spanish MEC under the MIPRCV “Consolider Ingenio 2010” research programme (CSD2007-00018), the iTransDoc research project (TIN2006-15694-CO2-01), and the FPU grant AP2005-1840.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Character-level interaction in multimodal computer-assisted transcription of text images

  • Martín-Albo Simón, Daniel
  • Romero Gómez, Verónica
  • Toselli, Alejandro Héctor|||0000-0001-6955-9249
  • Vidal, Enrique|||0000-0003-4579-5196
To date, automatic handwriting text recognition systems are far from being perfect and heavy human intervention is often required to check and correct the results of such systems. As an alternative, an interactive framework that integrates the human knowledge into the transcription process has been presented in previous works. In this work, multimodal interaction at character-level is studied. Until now, multimodal interaction had been studied only at whole-word level. However, character-level pen-stroke interactions may lead to more ergonomic and friendly interfaces. Empirical tests show that this approach can save significant amounts of user effort with respect to both fully manual transcription and non-interactive post-editing correction., Work supported by the Spanish Government (MICINN and “Plan E”) under the MITTRAL (TIN2009-14633-C03-01) research project and under the research programme Consolider Ingenio 2010: MIPRCV (CSD2007-00018), and by the Generalitat Valenciana under grant Prometeo/2009/014.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Discriminative Bernoulli Mixture Models for Handwritten Digit Recognition

  • Giménez Pastor, Adrián
  • Juan, Alfons|||0000-0002-9984-4072
  • Andrés Ferrer, Jesús
  • Serrano Martinez Santos, Nicolas
Bernoulli-based models such as Bernoulli mixtures or Bernoulli HMMs (BHMMs), have been successfully applied to several handwritten text recognition (HTR) tasks which range from character recognition to continuous and isolated handwritten words. All these models belong to the generative model family and, hence, are usually trained by (joint) maximum likelihood estimation (MLE). Despite the good properties of the MLE criterion, there are better training criteria such as maximum mutual information (MMI). The MMI is a widespread criterion that is mainly employed to train discriminative models such as log-linear (or maximum entropy) models. Inspired by the Bernoulli mixture classifier, in this work a log-linear model for binary data is proposed, the so-called mixture of multiclass logistic regression. The proposed model is proved to be equivalent to the Bernoulli mixture classifier. In this way, we give a discriminative training framework for Bernoulli mixture models. The proposed discriminative training framework is applied to a well-known Indian digit recognition task., Work supported by the EC (FEDER/FSE) and the Spanish MEC/MICINN under the MIPRCV “Consolider Ingenio 2010” program (CSD2007-00018), iTrans2 (TIN2009-14511) and MITTRAL (TIN2009-14633-C03-01) projects. Also supported by the IST Programme of the European Community, under the PASCAL2 Network of Excellence, IST-2007-216886, and by the Spanish MITyC under the erudito.com (TSI-020110-2009-439).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Online learning via dynamic reranking for Computer Assisted Translation

  • Martínez Gómez, Pascual
  • Sanchis Trilles, Germán
  • Casacuberta Nolla, Francisco|||0000-0002-8497-5598
New techniques for online adaptation in computer assisted translation are explored and compared to previously existing approaches. Under the online adaptation paradigm, the translation system needs to adapt itself to real-world changing scenarios, where training and tuning may only take place once, when the system is set-up for the first time. For this purpose, post-edit information, as described by a given quality measure, is used as valuable feedback within a dynamic reranking algorithm. Two possible approaches are presented and evaluated. The first one relies on the well-known perceptron algorithm, whereas the second one is a novel approach using the Ridge regression in order to compute the optimum scaling factors within a state-of-the-art SMT system. Experimental results show that such algorithms are able to improve translation quality by learning from the errors produced by the system on a sentence-by-sentence basis., This paper is based upon work supported by the EC (FEDER/FSE) and the Spanish MICINN under projects MIPRCV “Consolider Ingenio 2010” (CSD2007-00018) and iTrans2 (TIN2009-14511). Also supported by the Spanish MITyC under the erudito.com (TSI-020110-2009-439) project, by the Generalitat Valenciana under grant Prometeo/2009/014 and scholarship GV/2010/067 and by the UPV under grant 20091027
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

Interactive structured output prediction: Application to chromosome classification

  • Oncina Carratalá, José
  • Vidal, Enrique|||0000-0003-4579-5196
Interactive Pattern Recognition concepts and techniques are applied to problems with structured output; i.e., problems in which the result is not just a simple class label, but a suitable structure of labels. For illustration purposes (a simplification of) the problem of Human Karyotyping is considered. Results show that a) taking into account label dependencies in a karyogram significantly reduces the classical (non-interactive) chromosome label prediction error rate and b) they are further improved when interactive processing is adopted. c Springer-Verlag Berlin Heidelberg 2011, Work supported by the MIPRCV Spanish MICINN “Consolider Ingenio 2010” program (CSD2007-00018). Work supported by the Spanish CICyT through project TIN2009-14205-C04-01. This work was supported in part by the IST Programme of the European Community, under the PASCAL2 Network of Excellence, IST-2007- 216886.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Capítulo o parte de un libro (BookPart)

A Web-Based Demo to Interactive Multimodal Transcription of Historic Text images

  • Romero Gómez, Verónica
  • Leiva Torres, Luis Alberto
  • Alabau Gonzalvo, Vicente
  • Toselli, Alejandro Héctor|||0000-0001-6955-9249
  • Vidal, Enrique|||0000-0003-4579-5196
[EN] Paleography experts spend many hours transcribing historic documents, and state-of-the-art handwritten text recognition systems are not suitable for performing this task automatically. In this paper we present the modifications oil a previously developed interactive framework for transcription of handwritten text. This system, rather than full automation, aimed at assisting the user with the recognition-transcription process., This work has been supported by the EC (FEDER), the Spanish MEC under grant TIN2006-15694-C02-01 and the research programme Consolider Ingenio 2010 MIPRCV (CSD2007-00018) and by the UPV (FPI fellowship 2006-04).
Dipòsit Digital de Documents de la UAB
Publicación Capítulo o parte de un libro (BookPart)

Coloresia, An Interactive Colour Perception Device for the Visually Impaired

  • Gonzalez García, Abel
  • Benavente i Vidal, Robert|||0000-0001-9819-4445
  • Penacchio, Olivier|||0000-0002-1544-2405
  • Vazquez-Corral, Javier|||0000-0003-0414-7096
  • Vanrell i Martorell, Maria Isabel|||0000-0002-1567-9293
  • Parraga, Carlos Alejandro|||0000-0002-3809-241X
A significative percentage of the human population suffer from impairments in their capacity to distinguish or even see colours. For them, everyday tasks like navigating through a train or metro network map becomes demanding. We present a novel technique for extracting colour information from everyday natural stimuli and presenting it to visually impaired users as pleasant, non-invasive sound. This technique was implemented inside a Personal Digital Assistant (PDA) portable device. In this implementation, colour information is extracted from the input image and categorised according to how human observers segment the colour space. This information is subsequently converted into sound and sent to the user via speakers or headphones. In the original implementation, it is possible for the user to send its feedback to reconfigure the system, however several features such as these were not implemented because the current technology is limited.We are confident that the full implementation will be possible in the near future as PDA technology improves.
Dipòsit Digital de Documents de la UAB
Publicación Capítulo o parte de un libro (BookPart)

Saliency estimation using a non-parametric low-level vision model

  • Murray, Naila
  • Vanrell i Martorell, Maria Isabel|||0000-0002-1567-9293
  • Otazu Porter, Xavier|||0000-0002-4982-791X
  • Parraga, Carlos Alejandro|||0000-0002-3809-241X
Many successful models for predicting attention in a scene involve three main steps: convolution with a set of filters, a center-surround mechanism and spatial pooling to construct a saliency map. However, integrating spatial information and justifying the choice of various parameter values remain open problems. In this paper we show that an efficient model of color appearance in human vision, which contains a principled selection of parameters as well as an innate spatial pooling mechanism, can be generalized to obtain a saliency model that outperforms state-of-the-art models. Scale integration is achieved by an inverse wavelet transform over the set of scale-weighted center-surround responses. The scale-weighting function (termed ECSF) has been optimized to better replicate psychophysical data on color appearance, and the appropriate sizes of the center-surround inhibition windows have been determined by training a Gaussian Mixture Model on eye-fixation data, thus avoiding ad-hoc parameter selection. Additionally, we conclude that the extension of a color appearance model to saliency estimation adds to the evidence for a common low-level visual front-end for different visual tasks.
Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
Publicación Capítulo o parte de un libro (BookPart)

The impact of exact probabilistic learning algorithms in EDAs based on bayesian networks

  • Echegoyen Arruti, Carlos
  • Santana, Roberto
  • Lozano, José Antonio
  • Larrañaga, Pedro
This paper discusses exact learning of Bayesian networks in estimation of distribution algorithms. The estimation of Bayesian network algorithm (EBNA) is used to analyze the impact of learning the optimal (exact) structure in the search. By applying recently introduced methods that allow learning optimal Bayesian networks, we investigate two important issues in EDAs. First, we analyze the question of whether learning more accurate (exact) models of the dependencies implies a better performance of EDAs. Secondly, we are able to study the way in which the problem structure is translated into the probabilistic model when exact learning is accomplished. The results obtained reveal that the quality of the problem information captured by the probability model can improve when the accuracy of the learning algorithm employed is increased. However, improvements in model accuracy do not always imply a more efficient search., This work has been partially supported by the Etortek, Saiotek and Research Groups 2007-2012 (IT-242-07) programs (Basque Government), TIN2005-03824 and Consolider Ingenio 2010 - CSD2007-00018 projects (Spanish Ministry of Education and Science) and COMBIOMED network in computational biomedicine (Carlos III Health Institute).
Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
Publicación Capítulo o parte de un libro (BookPart)

Analyzing the k most probable solutions in EDAs based on bayesian networks

  • Echegoyen Arruti, Carlos
  • Mendiburu, Alexander
  • Santana, Roberto
  • Lozano, José Antonio
Acceso cerrado a este documento. No se encuentra disponible para la consulta pública. Depositado en Academica-e para cumplir con los requisitos de evaluación y acreditación académica del autor/a (sexenios, acreditaciones, etc.)., Estimation of distribution algorithms (EDAs) have been successfully applied to a wide variety of problems but, for themost complex approaches, there is no clear understanding of the way these algorithms complete the search. For that reason, in this work we exploit the probabilistic models that EDAs based on Bayesian networks are able to learn in order to provide new information about their behavior. Particularly, we analyze the k solutions with the highest probability in the distributions estimated during the search. In order to study the relationship between the probabilistic model and the fitness function, we focus on calculating, for the k most probable solutions (MPSs), the probability values, the function values and the correlation between both sets of values at each step of the algorithm. Furthermore, the objective functions of the k MPSs are contrasted with the k best individuals in the population. We complete the analysis by calculating the position of the optimum in the k MPSs during the search and the genotypic diversity of these solutions. We carry out the analysis by optimizing functions of different natures such as Trap5, two variants of Ising spin glass and Max-SAT. The results not only show information about the relationship between the probabilistic model and the fitness function, but also allow us to observe characteristics of the search space, the quality of the setup of the parameters and even distinguish between successful and unsuccessful runs., This work has been partially supported by the Saiotek and Research Groups 2007-2012 (IT-242-07) programs (Basque Government), TIN2008-06815-C02-01 and Consolider Ingenio 2010 - CSD2007-00018 projects (Spanish Ministry of Science and Innovation) and COMBIOMED network in computational biomedicine (Carlos III Health Institute). Carlos Echegoyen has a grant from UPV-EHU.