http://hdl.handle.net/2117/3778 2013-05-24T18:21:50Z http://hdl.handle.net/2117/19025 Title: The parameters of Menzerath-Altmann law in genomes Authors: Baixeries i Juvillà, Jaume; Hernández Fernández, Antonio; Forns, Núria; Ferrer Cancho, Ramon Abstract: The relationship between the size of the whole and the size of the parts in language and music is known to follow the Menzerath-Altmann law at many levels of description (morphemes, words, sentences, …). Qualitatively, the law states that the larger the whole, the smaller its parts, e.g. the longer a word (in syllables) the shorter its syllables (in letters or phonemes). This patterning has also been found in genomes: the longer a genome (in chromosomes), the shorter its chromosomes (in base pairs). However, it has been argued recently that mean chromosome length is trivially a pure power function of chromosome number with an exponent of -1. The functional dependency between mean chromosome size and chromosome number in groups of organisms from three different kingdoms is studied. The fit of a pure power function yields exponents between -1.6 and 0.1. It is shown that an exponent of -1 is unlikely for fungi, gymnosperm plants, insects, reptiles, ray-finned fishes and amphibians. Even when the exponent is very close to -1, adding an exponential component is able to yield a better fit with regard to a pure power-law in plants, mammals, ray-finned fishes and amphibians. The parameters of the Menzerath-Altmann law in genomes deviate significantly from a power law with a -1 exponent with the exception of birds and cartilaginous fishes. 2013-04-26T18:45:28Z http://hdl.handle.net/2117/17511 Title: The use of domain ontologies for improving the adaptability and collaborative ability of a web dialogue system Authors: González Bermúdez, Meritxell; Gatius Vila, Marta Abstract: Dialogue systems can be used for guiding the users accessing web services, enhancing the web usability. However, they are expensive to develop and difficult to adapt to different types of web services. The knowledge model of a web service can be seen as the basis to define the semantics of the information to be exchanged among the components of a dialogue system. This approach facilitates the integration of the different types of knowledge involved in human-machine communication and provides a unified framework easier to apply to new web services. Furthermore, the representation of the web service knowledge according to an ontology can enhance the reasoning capabilities of the underlying system. This article describes the use of domain ontologies in a mixed-initiative web dialogue system for improving both its adaptability and its collaborative ability. 2013-01-24T13:39:27Z http://hdl.handle.net/2117/17486 Title: Sibyl, a factoid question answering system for spoken documents Authors: Comas Umbert, Pere Ramon; Turmo Borras, Jorge; Màrquez Villodre, Lluís Abstract: In this article, we present a factoid question-answering system, Sibyl, specifically tailored for question answering (QA) on spoken-word documents. This work explores, for the first time, which techniques can be robustly adapted from the usual QA on written documents to the more difficult spoken document scenario. More specifically, we study new information retrieval (IR) techniques designed or speech, and utilize several levels of linguistic information for the speech-based QA task. These include named-entity detection with phonetic information, syntactic parsing applied to speech transcripts, and the use of coreference resolution. Sibyl is largely based on supervised machine-learning techniques, with special focus on the answer extraction step, and makes little use of handcrafted knowledge. Consequently, it should be easily adaptable to other domains and languages. Sibyl and all its modules are extensively evaluated on the European Parliament Plenary Sessions English corpus, comparing manual with automatic transcripts obtained by three different automatic speech recognition (ASR) systems that exhibit significantly different word error rates. This data belongs to the CLEF 2009 track for QA on speech transcripts. The main results confirm that syntactic information is very useful for learning to rank question candidates, improving results on both manual and automatic transcripts, unless the ASR quality is very low. At the same time, our experiments on coreference resolution reveal that the state-of-the-art technology is not mature enough to be effectively exploited for QA with spoken documents. Overall, the performance of Sibyl is comparable or better than the state-of-the-art on this corpus, confirming the validity of our approach. 2013-01-23T10:01:13Z http://hdl.handle.net/2117/16727 Title: A hybrid approach to treebank construction Authors: Marimon, Montserrat; Padró, Lluís Abstract: Este artículo describe investigación sobre los efectos de la desambiguación morfosintáctica usada como un preproceso de un analizador sintáctico profundo basado en HPSG, en el contexto del desarrollo de un treebank del español de código abierto, en el entorno de DELPH-IN. La anotación treebank se realiza manualmente tomando las decisiones apropiadas entre las opciones propuestas por el sistema y ordenadas por un módulo estadístico. Los experimentos presentados muestran que el uso de un etiquetador reduce la ambigüedad de las frases, y contribuye a limitar la cantidad de frases cuyo análisis sobrepasa el límite de tiempo, y ayuda a al módulo estadístico a clasificar el árbol correcto entre los mejores. Por un lado, nuestros resultados validan los beneficios ya reportados en la literatura de tal preproceso de análisis profundo con respecto a la velocidad, cobertura y precisión. Por otro lado, proponemos una estrategia basada en existentes herramientas de código abierto y recursos para desarrollar con alta consitencia treebanks de sintaxis profunda para idiomas con limitada disponibilidad de recursos lingüísticos. 2012-10-16T09:42:39Z http://hdl.handle.net/2117/14772 Title: Analizadores Multilingües en FreeLing Authors: Padró, Lluís Abstract: FreeLing es una librería de código abierto para el procesamiento multilíngüe automático, que proporciona una amplia gama de servicios de análisis lingüístico para diversos idiomas. FreeLing ofrece a los desarrolladores de aplicaciones de Procesamiento del Lenguaje Natural funciones de análisis y anotación lingüística de textos, con la consiguiente reducción del coste de construcción de dichas aplicaciones. FreeLing es personalizable y ampliable, y está fuertemente orientado a aplicaciones del mundo real en términos de velocidad y robustez. Los desarrolladores pueden utilizar los recursos lingüísticos por defecto (diccionarios, lexicones, gramáticas, etc), ampliarlos, adaptarlos a dominios particulares, o –dado que la librería es de código abierto– desarrollar otros nuevos para idiomas específicos o necesidades especiales de las aplicaciones. Este artículo presenta los principales cambios y mejoras incluidos en la versión 3.0 de FreeLing, y resume algunos proyectos industriales relevantes en los que se ha utilizado 2012-01-24T12:44:21Z http://hdl.handle.net/2117/14671 Title: Enhanced plant fault diagnosis based on the characterization of transient stages Authors: Monroy, Isaac; Benítez Iglesias, Raúl; Escudero Bakx, Gerard; Graells Sobré, Moisès Abstract: This paper introduces a data-based fault diagnosis system that includes an enhanced characterization of faults during transient stages. First, data under abnormal operating conditions (AOC) is projected onto a reference PCA model constructed with data under normal operating conditions (NOC). T2 and Q-statistic measures of this first PCA model are both used to detect the fault and to estimate the duration and delay of its transient evolution. After a dimensionality reduction, a second NOC PCA model is used to process data before diagnosing the faults by standard classification methods such as Artificial Neural Networks (ANN) or Support Vector Machines (SVM). A quantitative validation of the procedure has been carried out using simulated on-line data sets of the Tennessee Eastman Process (TEP). Results indicate that the incorporation of transient data in models improves the overall diagnosis performance, regardless of the particular choice between the statistical methods or the classification methods. 2012-01-19T12:10:25Z http://hdl.handle.net/2117/14563 Title: Random models of Menzerath-Altmann law in genomes Authors: Baixeries i Juvillà, Jaume; Hernández Fernández, Antonio; Ferrer Cancho, Ramon Abstract: Recently, a random breakage model has been proposed to explain the negative correlation between mean chromosome length and chromosome number that is found in many groups of species and is consistent with Menzerath–Altmann law, a statistical law that defines the dependency between the mean size of the whole and the number of parts in quantitative linguistics. Here, the central assumption of the model, namely that genome size is independent from chromosome number is reviewed. This assumption is shown to be unrealistic from the perspective of chromosome structure and the statistical analysis of real genomes. A general class of random models, including that random breakage model, is analyzed. For any model within this class, a power law with an exponent of −1 is predicted for the expectation of the mean chromosome size as a function of chromosome length, a functional dependency that is not supported by real genomes. The random breakage and variants keeping genome size and chromosome number independent raise no serious objection to the relevance of correlations consistent with Menzerath–Altmann law across taxonomic groups and the possibility of a connection between human language and genomes through that law. 2012-01-16T11:48:19Z http://hdl.handle.net/2117/13829 Title: Hacia la interacción en lenguaje natural Authors: Fuentes Fort, Maria; González Bermúdez, Meritxell Abstract: En éste documento se presenta la investigación que está siendo llevada a cabo en el Grupo de Procesamiento de Lenguaje Natural (GPLN) de la Universidad Politécnica de Cataluña (UPC). En concreto, hemos articulado la presentación de las diferentes líneas de trabajo tomando como referencia su aplicación en un asistente virtual. Creemos que su uso y implantación irá en aumento en los próximos diez años, de ahí la importancia del estado de las tecnologías del lenguaje natural y, aún mas, de los nuevos retos que este tipo de aplicaciones nos plantean. 2011-11-08T14:43:49Z http://hdl.handle.net/2117/13601 Title: ALICE: Acquisition of Language through an Interactive Comprehension Environment Authors: Fuentes Fort, Maria; González Bermúdez, Meritxell Abstract: Integration of several state-of-the-art technologies related to spoken language and natural language processing used in Intelligent Computer Assisted Language Learning (ICALL) systems. We envision to show that the technology has a level of maturity that suggests that the time may be right to use it at high school. // Integración de tecnologías del estado del arte en procesamiento del habla y procesamiento del lenguaje natural aplicadas a los asistentes inteligentes para el aprendizaje de lenguas. El objetivo es mostrar que el nivel de madurez de la tecnología permite que sea aplicada al aprendizaje de segundas lenguas en secundaria. 2011-10-20T12:06:46Z http://hdl.handle.net/2117/13368 Title: Size of the whole versus number of parts in genomes Authors: Hernández Fernández, Antonio; Baixeries i Juvillà, Jaume; Forns, Núria; Ferrer Cancho, Ramon Abstract: It is known that chromosome number tends to decrease as genome size increases in angiosperm plants. Here the relationship between number of parts (the chromosomes) and size of the whole (the genome) is studied for other groups of organisms from different kingdoms. Two major results are obtained. First, the finding of relationships of the kind "the more parts the smaller the whole" as in angiosperms, but also relationships of the kind "the more parts the larger the whole". Second, these dependencies are not linear in general. The implications of the dependencies between genome size and chromosome number are two-fold. First, they indicate that arguments against the relevance of the finding of negative correlations consistent with Menzerath-Altmann law (a linguistic law that relates the size of the parts with the size of the whole) in genomes are seriously flawed. Second, they unravel the weakness of a recent model of chromosome lengths based upon random breakage that assumes that chromosome number and genome size are independent. 2011-09-28T08:53:18Z http://hdl.handle.net/2117/13092 Title: Linguistic measures for automatic machine translation evaluation Authors: Giménez Linares, Jesús Ángel; Màrquez Villodre, Lluís Abstract: Assessing the quality of candidate translations involves diverse linguistic facets. However, most automatic evaluation methods in use today rely on limited quality assumptions, such as lexical similarity. This introduces a bias in the development cycle which in some cases has been reported to carry very negative consequences. In order to tackle this methodological problem, we explore a novel path towards heterogeneous automatic Machine Translation evaluation. We have compiled a rich set of specialized similarity measures operating at different linguistic dimensions and analyzed their individual and collective behaviour over a wide range of evaluation scenarios. Results show that measures based on syntactic and semantic information are able to provide more reliable system rankings than lexical measures, especially when the systems under evaluation are based on different paradigms. At the sentence level, while some linguistic measures perform better than most lexical measures, some others perform substantially worse, mainly due to parsing problems. Their scores are, however, suitable for combination, yielding a substantially improved evaluation quality. 2011-08-23T09:13:54Z http://hdl.handle.net/2117/10865 Title: The project HOPS: enabling an intelligent natural language based hub for the deployment of advanced semantically enriched multi-channel mass-scale online public services Authors: Gatius Vila, Marta; González Bermúdez, Meritxell Abstract: El objetivo principal del proyecto HOPS es desarrollar una plataforma multimodal y multilingüe que facilite el acceso de los ciutadanos a sus adminitraciones más próximas. El proyecto integra tecnologías de voz, procesamiento de lenguaje natural y web semántica. 2010-12-31T11:32:19Z http://hdl.handle.net/2117/10864 Title: Un sistema de diálogo multilingüe dirigido por la semántica Authors: Gatius Vila, Marta; González Bermúdez, Meritxell Abstract: Este artículo presenta un sistema de diálogos multilingüe basado en la semántica. El sistema utiliza una ontología que modela la aplicación para gestionar de forma eficiente la interacción oral y textual en diferentes lenguas (inglés, castellano y catalan). El conocimiento de la aplicación es utilizado por el gestor de diálogo para determinar la estructura del diálogo. También se utiliza para generar las gramáticas y léxicos en las diferentes lenguas. Estos recursos lingüísticos incorporan información de la aplicación para facilitar la interpretación semántica de las intervenciones del usuario. 2010-12-31T11:19:25Z http://hdl.handle.net/2117/10338 Title: Bases de conocimiento multilíngües para el procesamiento semántico a gran escala Authors: Cuadros Oller, Montserrat; Rigau Claramunt, German Abstract: Este artículo presenta el resultado del estudio de un amplio conjunto de bases de conocimiento multilingües actualmente disponibles que pueden ser de interés para un gran número de tareas de procesamiento semántico a gran escala. El estudio incluye una amplia gama de recursos derivados de forma manual y automática para el inglés y castellano. Con ello pretendemos mostrar una imagen clara de su estado actual. Para establecer una comparación justa y neutral, la calidad de cada recurso se ha evaluado indirectamente usando el mismo método en dos tareas de resolución de la ambigüedad semántica de las palabras (WSD, del inglés Word Sense Disambiguation). En concreto, las tareas de muestra léxica del ingles del Senseval-3. --- This report presents a wide survey of publicly available multilingual Knowledge Resources that could be of interest for wide–coverage semantic processing tasks. We also include an empirical evaluation in a multilingual scenario of the relative quality of some of these large-scale knowledge resources. The study includes a wide range of manually and automatically derived large-scale knowledge resources for English and Spanish. In order to establish a fair and neutral comparison, the quality of each knowledge resource is indirectly evaluated using the same method on a Word Sense Disambiguation task (Senseval-3 English Lexical Sample Task). 2010-11-17T13:16:05Z http://hdl.handle.net/2117/10337 Title: Multilingual evaluation of KnowNet Authors: Cuadros Oller, Montserrat; Rigau Claramunt, German Abstract: Este artículo presenta un nuevo método totalmente automático de construcción de bases de conocimiento muy densas y precisas a partir de recursos semánticos preexistentes. Básicamente, el método usa un algoritmo de Interpretación Semántica de las palabras preciso y de amplia cobertura para asignar el sentido mas apropiado a grandes conjuntos de palabras de un mismo tópico que han sido obtenidas de la web. KnowNet, la base de conocimiento resultante que conecta grandes conjuntos de conceptos semánticamente relacionados es un paso importante hacia la adquisición automática de conocimiento a partir de corpus. De hecho, KnowNet es varias veces mas grande que cualquier otro recurso de conocimiento disponible que codifique relaciones entre sentidos, y el conocimiento que KnowNet contiene supera cualquier otro recurso cuando es empíricamente evaluado en un marco multilingüe común. This paper presents a new fully automatic method for building highly dense and accurate knowledge bases from existing semantic resources. Basically, the method uses a wide-coverage and accurate knowledge-based Word Sense Disambiguation Algorithm to assign the most appropriate senses to large sets of topically related words acquired from the web. KnowNet, the resulting knowledge-base which connects large sets of semantically-related concepts is a major step towards the autonomous acquisition of knowledge from raw corpora. In fact, KnowNet is several times larger than any available knowledge resource encoding relations between synsets, and the knowledge KnowNet contains outperform any other resource when is empirically evaluated in a common multilingual framework. 2010-11-17T12:55:51Z