Papers by Carolina Gallardo
Proceedings of the 11th International Conference on Web Information Systems and Technologies, 2015
This paper is a first contribution of a schema to match professional and work competences through... more This paper is a first contribution of a schema to match professional and work competences through similarity measures. In this contribution we focus on the determination of connections between university profiles based in standards (body of knowledge and thesauri) similarity measures and Natural Language Processing (NLP) techniques. Our first experiments proved that this hybrid schema got a promise results in the recognition of competency patterns in order to apply in the laboral context.
Language Resources and Evaluation, May 1, 2002
The UNL System was conceived to support multilingual services in Internet being an alternative to... more The UNL System was conceived to support multilingual services in Internet being an alternative to the classical machine translation systems. The UNL System is based in the creation of documents written in an unique computational language able to represent concepts and their relations. The definition of this language has been possible thanks to the collaboration of more than one hundred people, prestigious researchers, and scientists of all around the world. The purpose of the UNL is to break the linguistic barriers in Internet that avoid the real global access to the knowledge and culture for all people. Since the starting of the UNL project in 1996, the participants in the project from initially 14 languages have made substantial progress in the technical as well as the organizational aspects involved. This demonstration will show to the academic and business communities the current state and practical achievements of the UNL system.
Springer eBooks, 2011
This article describes the work performed over the datábase of questions belonging to the differe... more This article describes the work performed over the datábase of questions belonging to the different opinión polis carried during the last 50 years in Spain. Approximately half of the questions are provided with a title while the other half remain untitled. The work and implemented techniques in order to automatically genérate the titles for untitled questions are described. This process is performed over very short texts and generated titles are subject to strong stylistic conventions and should be fully grammatical pieces of Spanish.
This paper presents the innovations in the practical work of the Data Structures subject carried ... more This paper presents the innovations in the practical work of the Data Structures subject carried out in the last five years, including a transition period and a first year of implantation of the European Higher Education Area. The practical coursework is inspired by a project-based methodology and from 2008/2009 additional laboratory sessions are included in the subject schedule. We will present the academic results and ratios of the mentioned time period which imply a significant improvement on students' performance.
A new e-learning tool for cognitive democracies in the Knowledge Society
Computers in Human Behavior, 2014
ABSTRACT Cognitive democracies can be defined as new models of e-democracy that are based on the ... more ABSTRACT Cognitive democracies can be defined as new models of e-democracy that are based on the evolutionism of living systems and deal with their vital (cognitive) process by means of the continuous education of citizens with regards to one of the essential characteristics of humankind: decision making. The education is provided by the social democratisation of the knowledge. This corresponds to the proportion of the arguments that support the positions. These arguments are extracted from the opinions expressed in the discussion stages included in the e-democracy models. (Moreno-Jiménez et al., 2012) presented a collaborative platform which is currently being used as the methodological support for the cognitive democracy known as e-cognocracy (Moreno-Jiménez, 2003a, Moreno-Jiménez, 2004, Moreno-Jiménez, 2006, Moreno-Jiménez and Polasek, 2003 and Moreno-Jiménez and Polasek, 2005) and for identifying and classifying the messages or opinions that favour the different individual positions through the employment of quantitative (data mining) and qualitative (text mining) approaches. This paper represents the next step in the phase concerned with extraction of the policy-making arguments. Using opinion analysis techniques, the new e-learning tool extracts the arguments that support the different opinions. This tool has been applied to the discussion stage of a case study developed at the University of Zaragoza concerning the possible location of Europe’s biggest leisure complex (Gran Scala) in Aragon (Spain).
Resumen-El Espacio Europeo de Educación Superior (EEES) centra la metodología de aprendizaje en e... more Resumen-El Espacio Europeo de Educación Superior (EEES) centra la metodología de aprendizaje en el propio alumno, quien con la ayuda del profesor debe pasar a ser agente activo, protagonizando su propio aprendizaje. La acción tutorial desempeña un papel fundamental para catalizar esta ayuda facilitando una comunicación fluida entre ambos. El objetivo de este trabajo es determinar la influencia que la utilización de herramientas TIC para la reserva y seguimiento de tutorías puede ejercer en el proceso de aprendizaje activo de los alumnos. A tal fin se ha llevado a cabo un estudio empírico de carácter descriptivo-exploratorio centrado en la
Modelo computable enriquecido de adquisición y representación de conocimiento lingüístico para procesamiento multidireccional de la lengua española
La aproximacion mas tradicional a la formalizacion y tratamiento del conocimiento linguistico se ... more La aproximacion mas tradicional a la formalizacion y tratamiento del conocimiento linguistico se ha basado en las gramaticas formales derivadas de teorias linguisticas. Sin embargo, las necesidades de sistemas de PLN para aplicaciones reales han descubierto las deficiencias que soportan los sistemas basados en teorias linguisticas. Estas son aptas para crear herramientas de investigacion teorica, pero inadecuadas para el diseno de sistemas orientados a la aplicacion real, debido a la excesiva rigidez que el marco teorico de una teoria linguistica impone sobre el conocimiento linguistico. Las teorias linguisticas no son el unico "conocimiento" que la linguistica proporciona. Las gramaticas descriptivas, aunque no estan orientadas a un uso computacional, ofrecen una descripcion del uso de la lengua provista de una mayor cobertura de conocimiento gramatical a pesar de su menor grado de formalizacion. Debido a este menor grado de formalizacion, el PLN ha dirigido sus trabajos a los modelos mas directamente computables, los derivados de las gramaticas teoricas en detrimento de gramaticas de tipo descriptivo mucho mas orientadas a los datos obtenidos por observacion que a la aplicacion computacional, de manera que el cumulo de conocimiento gramatical contenido en una gramatica descriptiva queda obviado. En este contexto, se ha orientado este trabajo tomando como objetivo la definicion de un modelo que permita educir el conocimiento de la lengua espanola contenido en la mayor obra de referencia hasta el momento: la "Gramatica Descriptiva de la Lengua Espanola" (GDLE) de la Real Academia Espanola, y representarlo de una manera computable de forma independiente de su aplicacion final. Para ello se ha realizado un analisis profundo desde la fuente (GDLE) -mediante el uso de metodologias de ingenieria del conocimiento- modelizando sus contenidos por medio de una representacion de conocimiento b
This paper describes the followed methodology to automatically generate titles for a corpus of qu... more This paper describes the followed methodology to automatically generate titles for a corpus of questions that belong to sociological opinion polls. Titles for questions have a twofold function: (1) they are the input of user searches and (2) they inform about the whole contents of the question and possible answer options. Thus, generation of titles can be considered as a case of automatic summarization. However, the fact that summarization had to be performed over very short texts together with the aforementioned quality conditions imposed on new generated titles led the authors to follow knowledge-rich and domain-dependent strategies for summarization, disregarding the more frequent extractive techniques for summarization.
Learning to Develop Software with Flipped Classroom
Advanced Science Letters, Apr 1, 2018
«Simple extractor»: una aplicación informática sencilla para la extracción y gestión terminológica
SimpleExtractor es una aplicacion informatica orientada a la extraccion de terminologia a partir ... more SimpleExtractor es una aplicacion informatica orientada a la extraccion de terminologia a partir de textos. Una de sus principales caracteristicas es su sencillez de manejo y sus interfaces intuitivas. La herramienta permite configurar aspectos de la extraccion y exportar a ficheros.
Procesamiento Del Lenguaje Natural, Sep 1, 2002
The National Institute for Statistics is the organism responsible for acquiring economical data f... more The National Institute for Statistics is the organism responsible for acquiring economical data for governmental statistics purposes. Lisbon agreements establish a framework in which this acquisition process shall be available through Internet, so each survey should be considered as a little software project to be developed and maintained. Considering the great amount of different surveys and all changes produced per year on each make impossible this task. An application generator has been developed to automate this task, taking as a start point the Word or PDF template of a survey, and going through a graphical form designer as all human effort, all HTML, Java classes and Oracle database resources are generated and sent from backoffice to frontoffice servers, reducing the team to carry out the whole set of electronic surveys to two people from non I.T. staff.
QoS for JPEG2000 Storage System: Based on Data Structure
2013 IEEE 27th International Conference on Advanced Information Networking and Applications (AINA), 2013
Quality of Service (QoS) is crucial for applications like professional video production, where us... more Quality of Service (QoS) is crucial for applications like professional video production, where users have to simultaneously access to huge multimedia files in a specific period of time. Different tasks have different requirements, some of them need high bandwidth while others require an error-free environment. Video production companies have to choose between the use of a specific storage system for each group of tasks and the use of a central storage system that meets more stringent requirements. In this paper, we propose a storage system that combines multiple disk drive components. The video file is distributed across the drives, depending on the level of redundancy and performance that is required in each part of the data. The number of simultaneous users indicates the performance of individual disks and the significance of the data specifies the redundancy. With this architecture it is possible to adapt the central storage system to individual needs and thus saves cost.
This paper describes the followed methodology to automatically generate titles for a corpus of qu... more This paper describes the followed methodology to automatically generate titles for a corpus of questions that belong to sociological opinion polls. Titles for questions have a twofold function: (1) they are the input of user searches and (2) they inform about the whole contents of the question and possible answer options. Thus, generation of titles can be considered as a case of automatic summarization. However, the fact that summarization had to be performed over very short texts together with the aforementioned quality conditions imposed on new generated titles led the authors to follow knowledge-rich and domain-dependent strategies for summarization, disregarding the more frequent extractive techniques for summarization. Management information based on summaries is frequent in libraries, documentation centers, press or document collections based on short texts. Some libraries have at their disposal information based on summaries of articles or abstracts that have been not generat...
The UNL System was conceived to support multilingual services in Internet being an alternative to... more The UNL System was conceived to support multilingual services in Internet being an alternative to the classical machine translation systems. The UNL System is based in the creation of documents written in an unique computational language able to represent concepts and their relations. The definition of this language has been possible thanks to the collaboration of more than one hundred people, prestigious researchers, and scientists of all around the world. The purpose of the UNL is to break the linguistic barriers in Internet that avoid the real global access to the knowledge and culture for all people. Since the starting of the UNL project in 1996, the participants in the project from initially 14 languages have made substantial progress in the technical as well as the organizational aspects involved. This demonstration will show to the academic and business communities the current state and practical achievements
«Simple extractor»: una aplicación informática sencilla para la extracción y gestión terminológica
SimpleExtractor es una aplicacion informatica orientada a la extraccion de terminologia a partir ... more SimpleExtractor es una aplicacion informatica orientada a la extraccion de terminologia a partir de textos. Una de sus principales caracteristicas es su sencillez de manejo y sus interfaces intuitivas. La herramienta permite configurar aspectos de la extraccion y exportar a ficheros.
Learning to Develop Software with Flipped Classroom
Advanced Science Letters, 2018
Revista española de Documentación Científica, 2016
Resumen: Este artículo describe el trabajo realizado para la generación automática de los títulos... more Resumen: Este artículo describe el trabajo realizado para la generación automática de los títulos de las preguntas pertenecientes a las encuestas de opinión que existen en las bases de datos del CIS (Centro de Investigaciones Sociológicas). Dentro del contexto del CIS, el título de una pregunta debe cumplir dos requisitos: desde el punto de vista de la forma, debe ser gramaticalmente correcto y tener un estilo similar a los ya existentes; y, desde el punto de vista del contenido, debe albergar el tema de la pregunta y las distintas categorías de respuesta. Estas restricciones en cuanto a la forma y al contenido de los títulos desaconsejan el uso de técnicas empleadas en problemas similares, como el resumen automático o aprendizaje automático con corpus de entrenamiento, a favor de una metodología basada en el análisis y conocimiento del dominio. Para ilustrar el análisis y la estrategia de resolución del problema seguidos, hemos seleccionado las preguntas relacionadas con temas electorales, debido a la importancia estratégica y a la especialización del CIS en este tipo de encuestas. Se describe en detalle el procedimiento seguido y la evaluación de los resultados, valorando tanto los aspectos cualitativos como los cuantitativos. La evaluación muestra que el 88,73% de los títulos generados cumplen estrictamente con los requisitos de forma y contenido impuestos por el CIS, lo que supone un ahorro en el trabajo manual del personal cualificado de la institución. Palabras clave: Minería de textos; recuperación de información; filtrado; clasificación; resumen y visualización; titulación automática; extracción de información; encuestas de opinión.
Procesamiento Del Lenguaje Natural, 2002
The best results in the application of computer science systems to automatic translation are obta... more The best results in the application of computer science systems to automatic translation are obtained in word processing when texts pertain to specific thematic areas, with structures well defined and a concise and limited lexicon. In this article we present a plan of systematic work for the analysis and generation of language applied to the field of pharmaceutical leaflet, a type of document characterized by format rigidity and precision in the use of lexicon. We propose a solution based in the use of one interlingua as language pivot between source and target languages; we are considering Spanish and Arab languages in this case of application.
Uploads
Papers by Carolina Gallardo