Search

This introduction provides an overview of the state-of-the-art technology in Applications of Natural Language to Information Systems. Specifically, we analyze the need for such technologies to successfully address the new challenges of modern information systems, in which the exploitation of the Web as a main data source on business systems becomes a key requirement. It will also discuss the reasons why Human Language Technologies themselves have shifted their focus onto new areas of interest very directly linked to the development of technology for the treatment and understanding of Web 2.0. These new technologies are expected to be future interfaces for the new information systems to come. Moreover, we will review current topics of interest to this research community, and will present the selection of manuscripts that have been chosen by the program committee of the NLDB 2011 conference as representative cornerstone research works, especially highlighting their contribution to the advancement of such technologies.

In the chemical textile domain experts have to analyse chemical components and substances that might be harmful for their usage in clothing and textiles. Part of this analysis is performed searching opinions and reports people have expressed concerning these products in the Social Web. However, this type of information on the Internet is not as frequent for this domain as for others, so its detection and classification is difficult and time-consuming. Consequently, problems associated to the use of chemical substances in textiles may not be detected early enough, and could lead to health problems, such as allergies or burns. In this paper, we propose a framework able to detect, retrieve, and classify subjective sentences related to the chemical textile domain, that could be integrated into a wider health surveillance system. We also describe the creation of several datasets with opinions from this domain, the experiments performed using machine learning techniques and different lexical resources such as WordNet, and the evaluation focusing on the sentiment classification, and complaint detection (i.e., negativity). Despite the challenges involved in this domain, our approach obtains promising results with an F-score of 65% for polarity classification and 82% for complaint detection.

El objetivo de este proyecto se basa en la necesidad de replantearse la filosofía clásica del TLH para adecuarse tanto a las fuentes disponibles actualmente (datos no estructurados con multi-modalidad, multi-lingualidad y diferentes grados de formalidad) como a las necesidades reales de los usuarios finales. Para conseguir este objetivo es necesario integrar tanto la comprensión como la generación del lenguaje humano en un modelo único (modelo LEGOLANG) basado en técnicas de deconstrucción de la lengua, independiente de su aplicación final y de la variante de lenguaje humano elegida para expresar el conocimiento., The main objective of this project is based on the need to reconsider the clasical HLT philosophy to adapt it, not only to the currently available resources (unstructured data with multimodality, multilinguality and different levels of formality) but also to the real needs of the final users. In order to reach this objective it is necessary to include the understanding as well as the generation of human language in a unique model (LEGOLANG model) based on language deconstruction techniques, independently of the final application and the human language variant chosen to express the knowledge.

Ejercicios basados en los esquemas Ejemplo y Tienda On Line de la práctica SQL de la asignatura Fundamentos de las Bases de Datos.

Introducción a SQL.

Exámenes solucionados de segunda convocatoria de 2011.

Mapas conceptuales de los temas de la asignatura.

Infografía mostrando estadísticas sobre los resultados de las dos convocatorias de Fundamentos de las Bases de Datos en el curso 2011-12. Comentado en http://blogfbd.blogspot.com.es/2012/08/resultados-del-curso-2011-12.html

La segunda hora de las sesiones de teoría se centra en el modelo relacional y tiene un carácter más práctico, de resolución de problemas de aplicación de lo anterior.

Presentaciones de la primera hora de las sesiones de teoría (TEORÍA1).

Examen de segunda convocatoria del curso 2011-12. Consta de 3 partes: test y problemas de teoría, y SQL.

Ejercicios sobre los conceptos desarrollados en la primera parte de las sesiones de teoría (hora1, teoría1).

Extracto de los apuntes referido al cálculo relacional y la perspectiva del modelo relacional desde el cálculo de predicados de primer orden.

Tesis doctoral con mención europea en procesamiento del lenguaje natural realizada en la Universidad de Alicante por Ester Boldrini bajo la dirección del Dr. Patricio Martínez-Barco. El acto de defensa de la tesis tuvo lugar en la Universidad de Alicante el 23 de enero de 2012 ante el tribunal formado por los doctores Manuel Palomar (Universidad de Alicante), Dr. Paloma Moreda (UA), Dr. Mariona Taulé (Universidad de Barcelona), Dr. Horacio Saggion (Universitat Pompeu Fabra) y Dr. Mike Thelwall (University of Wolverhampton). Calificación: Sobresaliente Cum Laude por unanimidad., European Ph.D Thesis in Computational Linguistics, written at the Univeridad de Alicante by Ester Boldrini under the supervision of Dr. Patricio Martínez-Barco. The author was examined on January, 23th 2012 at the Universidad de Alicante by a commission composed by Dr. Manuel Palomar (Universidad de Alicante), Dr. Paloma Moreda (UA), Dr. Mariona Taulé (Universidad de Barcelona), Dr. Horacio Saggion (Universitat Pompeu Fabra) and Dr. Mike Thelwall (University of Wolverhampton). The mark obtained was Sobresaliente Cum Laude.

The Internet boom in recent years has increased the interest in the field of plagiarism detection. A lot of documents are published on the Net everyday and anyone can access and plagiarize them. Of course, checking all cases of plagiarism manually is an unfeasible task. Therefore, it is necessary to create new systems that are able to automatically detect cases of plagiarism produced. In this paper, we introduce a new hybrid system for plagiarism detection which combines the advantages of the two main plagiarism detection techniques. This system consists of two analysis phases: the first phase uses an intrinsic detection technique which dismisses much of the text, and the second phase employs an external detection technique to identify the plagiarized text sections. With this combination we achieve a detection system which obtains accurate results and is also faster thanks to the prefiltering of the text.

Preliminary research demonstrated the EmotiBlog annotated corpus relevance as a Machine Learning resource to detect subjective data. In this paper we compare EmotiBlog with the JRC Quotes corpus in order to check the robustness of its annotation. We concentrate on its coarse-grained labels and carry out a deep Machine Learning experimentation also with the inclusion of lexical resources. The results obtained show a similarity with the ones obtained with the JRC Quotes corpus demonstrating the EmotiBlog validity as a resource for the SA task.

In this paper we present a method to automatically identify linguistic contexts which contain possible causes of emotions or emotional states from Italian newspaper articles (La Repubblica Corpus). Our methodology is based on the interplay between relevant linguistic patterns and an incremental repository of common sense knowledge on emotional states and emotion eliciting situations. Our approach has been evaluated with respect to manually annotated data. The results obtained so far are satisfying and support the validity of the methodology proposed.

EmotiBlog is a corpus designed for Sentiment Analysis research. Preliminary studies demonstrated its relevance as a Machine Learning resource for detecting subjective information. In this paper we explore additional features by a detailed analysis. In addition, we compare EmotiBlog with other well-known Sentiment Analysis resource such as the JRC corpus. Finally, as a result of our research, we developed an Opinion Mining application, which takes into account user opinions when rating the results of a search engine specialized in mobile phones.

Un sistema de recuperación de información orientado a la minería de opiniones, en el que se investiguen las influencias positivas entre las tres áreas que abarca: recuperación de información, rastreo web y minería de opiniones.

El objetivo general de este proyecto se centra en el estudio, desarrollo y experimentación de diferentes técnicas y sistemas basados en Tecnologías del Lenguaje Humano (TLH) para el desarrollo de la próxima generación de sistemas de procesamiento inteligente de la información digital (modelado, recuperación, tratamiento, comprensión y descubrimiento) afrontando los actuales retos de la comunicación digital. En este nuevo escenario, los sistemas deben incorporar capacidades de razonamiento que descubrirán la subjetividad de la información en todos sus contextos (espacial, temporal y emocional) analizando las diferentes dimensiones de uso (multilingualidad, multimodalidad y registro)., The overall aim of this project focuses on the study, development and experimentation with different techniques and systems based on Human Language Technologies (HLT) for developing the next generation of intelligent digital information processing systems (modelling, retrieval, processing, comprehension and detection), in order to meet the present challenges posed by digital media. In this new scenario, systems have to incorporate the reasoning capability to ascertain the subjectivity of information in all contexts (spatial, temporal and emotional), while analysing the various dimensional uses (multilingualism, multimodality and register).

El proyecto Araknion tiene como objetivo general dotar al español y al catalán de una infraestructura básica de recursos lingüísticos para el procesamiento semántico de corpus en el marco de la Web 2.0 sean de origen oral o escrito., Araknion project aims to provide the Spanish and Catalan with basic linguistic resources (oral or written) for semantic processing in the context of Web 2.0.

EmotiBlog es una colección de entradas de blogs creado y anotado para detectar expresiones subjetivas en los nuevos géneros textuales nacidos con la Web 2.0. Investigaciones previas han demostrado la relevancia de los sistemas de aprendizaje automático como recurso para la detección de información de opinión. En este artículo exploramos características adicionales para un análisis profundo de estas técnicas. Además, comparamos EmotiBlog con la colección JRC. Los resultados obtenidos demuestran la validez de EmotiBlog y nos animan a seguir en esta línea de investigación., EmotiBlog is a collection of blog posts created and annotated for detecting subjective expressions in the new textual genres born with the Web 2.0. Previous work has demonstrated the relevance of the Machine learning systems as tool for detecting opinionated information. In this paper we explore additional features for a deep analysis of these techniques. Moreover, we compare EmotiBlog with the JRC collection. The obtained results demonstrate the usefulness of EmotiBlog and support us to continue in this research path.

La detección de menciones es el primer módulo utilizado en los sistemas de resolución de la correferencia. Debido a que los errores se van acarreando de un módulo a otro, es importante que los resultados obtenidos por este módulo sean los mejores posibles. Dentro del campo de la detección de menciones, las menciones anidadas son las más difíciles de detectar. En este artículo presentamos un sistema de detección de menciones anidadas basado en expansión, un nuevo modelo de detección de elementos de Procesamiento de Lenguaje Natural anidados basado en aprendizaje automático. Los resultados obtenidos por nuestro sistema son superiores al 72 % de medida-F en el corpus AnCora. No podemos comparar directamente nuestros resultados con otros sistemas debido a que no existen como tales, pero si comparamos la medida-F media obtenida por otros sistemas en la detección de todo tipo de menciones (no solo menciones anidadas), y que nosotros estamos tratando con las menciones más difíciles, conseguimos resultados favorables., Mention detection is the first module used in coreference resolution systems. Due to that, it is important that the results obtained by this module are as high as possible. Within the field of mention detection, nested mentions are the most difficult ones to detect. In this paper, we present a nested mention detection system based on expansion, a new model for detecting nested elements in NLP based on machine learning. The results obtained by our system are above the 72 % in F-measure in AnCora corpus. We can not compare directly our results with other systems, since there are not exist, but if we consider that the average F-measure obtained by other systems for all mention (not only nested mentions), and that we are dealing with the most difficult mentions, we achieve good results.

EmotiBlog is a corpus labelled with the homonymous annotation schema designed for detecting subjectivity in the new textual genres. Preliminary research demonstrated its relevance as a Machine Learning resource to detect opinionated data. In this paper we compare EmotiBlog with the JRC corpus in order to check the EmotiBlog robustness of annotation. For this research we concentrate on its coarse-grained labels. We carry out a deep ML experimentation also with the inclusion of lexical resources. The results obtained show a similarity with the ones obtained with the JRC demonstrating the EmotiBlog validity as a resource for the SA task.

[ES] El objetivo general de este proyecto se centra en el estudio, desarrollo y experimentación de diferentes técnicas y sistemas basados en Tecnologías del Lenguaje Humano (TLH) para el desarrollo de la próxima generación de sistemas de procesamiento inteligente de la información digital (modelado, recuperación, tratamiento, comprensión y descubrimiento) afrontando los actuales retos de la comunicación digital. En este nuevo escenario, los sistemas deben incorporar capacidades de razonamiento que descubrirán la subjetividad de la información en todos sus contextos (espacial, temporal y emocional) analizando las diferentes dimensiones de uso (multilingualidad, multimodalidad y registro)., [EN] The overall aim of this project focuses on the study, development and experimentation with different techniques and systems based on Human Language Technologies (HLT) for developing the next generation of intelligent digital information processing systems (modelling, retrieval, processing, comprehension and detection), in order to meet the present challenges posed by digital media. In this new scenario, systems have to incorporate the reasoning capability to ascertain the subjectivity of information in all contexts (spatial, temporal and emotional), while analysing the various dimensional uses (multilingualism, multimodality and register).

Tradicionalmente, en tareas de recuperación de información, se han utilizado los métodos de evaluación definidos en foros internacionales como el TREC o el CLEF. Dichos métodos se basan, principalmente, en la cobertura y precisión a partir de los juicios de relevancia de las primeras n respuestas de cada sistema para un conjunto de consultas iniciales. En la práctica, cuando un buscador se aplica a la Web, los usuarios no quieren una alta cobertura sino una alta precisión para obtener páginas muy relevantes en las primeras posiciones. Es por ello que hemos adaptado los métodos de evaluación de estos foros valorando más la precisión que la cobertura. Finalmente, se ha puesto en práctica éste nuevo método de evaluación comparando dos buscadores web sobre un dominio restringido: oportunidades de negocio y exportación de productos., Traditionally, information retrieval systems are assessed using the same evaluation methods as proposed in the international forums, like TREC or CLEF. Such methods mainly focus on the recall and precision obtained from the relevance judgements of the first n answers each system provides for a set of initial queries. In practice, when applying a search engine to the Web, users do not want a high recall but a high precision in order to obtain really relevant pages at the first places. Therefore, in this paper we have adapted the evaluation methods of these forums so that precision is given more importance than recall. To test this new evaluation method, two web search engines are compared within the following restricted domains: business oportunities and product exports.

The increasing amount of subjective data on the Web is creating the need to develop effective Question Answering systems able to discriminate such information from factual data, and subsequently process it with specific methods. The participants in the IBEREVAL OM tasks will be given a set of opinion questions (in Spanish and English). Optionally, they will also be able to receive the same set of opinion questions, in which the source, target and expected polarity, as well as the time span the question is referring to are given. They will also be provided with a collection of blog posts, extracted using the Technorati blog search engine (in Spanish and English), in which the answers to the opinion questions should be found The gold standard for this blog posts collection will previously be annotated using the EmotiBlog scheme, by a number of 3 annotators. The EmotiBlog corpus and the set of questions presented in (Balahur et al., 2009) – in their present state will be provided for system training. The participants will be able to participate in two subtasks : 1) in the first one, they will be asked to provide the list of answers to each of the questions (in the same language as the questions, or in the other language); 2) in the second one, they will be asked to provide a summary of the question answers – the top x% of the most important answers, in a non-redundant manner. The Gold Standard for the summaries will be automatically extracted from the manual annotations, taking into account the “intensity” parameter of the opinions expressed., Con el grande aumento de la información subjetiva en la Web, hay una importante necesidad de desarrollar sistemas de Question Answering que sean eficientes y capaces de discriminar entre datos objetivos y subjetivos. Los participantes tendrán una colección de preguntas de opinión (Español e Inglés) en las cuales se deberán encontrar las respuestas. El Gold Standard será anotado previamente con el esquema de anotación EmotiBlog por 3 anotadores. El corpus EmotiBlog y la colección de preguntas presentados en (Balahur et al. 2009) se pondrá a disposición para el entrenamiento del sistema. Los participantes deberán devolver un listado de respuestas para cada una de las preguntas, (en el mismo idioma que la pregunta o en otro), un resumen de las respuestas –de las x% de las respuestas más importantes, de una manera no redundante, el Gold Standard para los resúmenes será extraído automáticamente de las anotaciones manuales teniendo en consideración el parámetro de “intensidad” de la opinión expresada.

The exponential growth of the subjective information in the framework of the Web 2.0 has led to the need to create Natural Language Processing tools able to analyse and process such data for multiple practical applications. They require training on specifically annotated corpora, whose level of detail must be fine enough to capture the phenomena involved. This paper presents EmotiBlog – a fine-grained annotation scheme for subjectivity. We show the manner in which it is built and demonstrate the benefits it brings to the systems using it for training, through the experiments we carried out on opinion mining and emotion detection. We employ corpora of different textual genres –a set of annotated reported speech extracted from news articles, the set of news titles annotated with polarity and emotion from the SemEval 2007 (Task 14) and ISEAR, a corpus of real-life self-expressed emotion. We also show how the model built from the EmotiBlog annotations can be enhanced with external resources. The results demonstrate that EmotiBlog, through its structure and annotation paradigm, offers high quality training data for systems dealing both with opinion mining, as well as emotion detection.

The present is marked by the availability of large volumes of heterogeneous data, whose management is extremely complex. While the treatment of factual data has been widely studied, the processing of subjective information still poses important challenges. This is especially true in tasks that combine Opinion Analysis with other challenges, such as the ones related to Question Answering. In this paper, we describe the different approaches we employed in the NTCIR 8 MOAT monolingual English (opinionatedness, relevance, answerness and polarity) and cross-lingual English-Chinese tasks, implemented in our OpAL system. The results obtained when using different settings of the system, as well as the error analysis performed after the competition, offered us some clear insights on the best combination of techniques, that balance between precision and recall. Contrary to our initial intuitions, we have also seen that the inclusion of specialized Natural Language Processing tools dealing with Temporality or Anaphora Resolution lowers the system performance, while the use of topic detection techniques using faceted search with Wikipedia and Latent Semantic Analysis leads to satisfactory system performance, both for the monolingual setting, as well as in a multilingual one.

The treatment of factual data has been widely studied in different areas of Natural Language Processing (NLP). However, processing subjective information still poses important challenges. This paper presents research aimed at assessing techniques that have been suggested as appropriate in the context of subjective - Opinion Question Answering (OQA). We evaluate the performance of an OQA with these new components and propose methods to optimally tackle the issues encountered. We assess the impact of including additional resources and processes with the purpose of improving the system performance on two distinct blog datasets. The improvements obtained for the different combination of tools are statistically significant. We thus conclude that the proposed approach is adequate for the OQA task, offering a good strategy to deal with opinionated questions.

En este documento se describe la planificación docente, los contenidos, la metodología y la evaluación de la asignatura Fundamentos Informáticos que se imparte en el Grado de Ingeniería de la Edificación de la Universidad de Alicante.

El crecimiento exponencial de la información subjetiva en el marco de la Web 2.0 ha creado la necesidad de producir herramientas de Procesamiento del Lenguaje Natural que sean capaces de analizar y procesar estos datos para aplicaciones concretas. Estas herramientas requieren un entrenamiento con corpus anotados con este tipo de información a nivel muy detallado para poder capturar aquellos fenómenos lingüísticos que contienen una carga emotiva. El presente artículo describe EmotiBlog, un modelo detallado para la anotación de la subjetividad. Presentamos el proceso de creación y demostramos que aporta mejoras a los sistemas de aprendizaje automático. Para ello, empleamos distintos corpus que presentan textos de diversos géneros – una colección de noticias periodísticas en estilo indirecto, la colección de títulos de noticias anotados con la polaridad y emoción del SemEval 2007 (Tarea 14) e ISEAR, un corpus de expresiones reales de emociones. Además, demostramos que otros recursos pueden integrarse con EmotiBlog. Los resultados prueban que gracias a su estructura y parámetros de anotación, el modelo propuesto, EmotiBlog, proporciona ventajas considerables para el entrenamiento de sistemas que trabajan con minería de opiniones y detección de emoción., The exponential growth of the subjective information in the framework of the Web 2.0 has led to the need to create Natural Language Processing tools able to analyse and process such data for multiple practical applications. These applications require training on specifically annotated corpora, whose level of detail must be fine enough to capture the phenomena involved. This paper presents EmotiBlog – a fine-grained annotation scheme for subjectivity. We show the manner in which it is built and demonstrate the benefits it brings to the systems using it for training, through the experiments we carried out on opinion mining and emotion detection. We employ corpora of different textual genres –a set of annotated reported speech extracted from news articles, the set of news titles annotated with polarity and emotion from the SemEval 2007 (Task 14) and ISEAR, a corpus of real-life self-expressed emotion. We also show how the model built from the EmotiBlog annotations can be enhanced with external resources. The results demonstrate that EmotiBlog, through its structure and annotation paradigm, offers high quality training data for systems dealing both with opinion mining, as well as emotion detection.