Textual Information Extraction

Material completo EIT

TEXT-MESS: Intelligent, Interactive and Multilingual Text Mining based on Human Language Technologies, TIN2006-15265-C06

The goal of the project is to analyze, experiment, and develop intelligent, interactive and multilingual Text Mining technologies, as a key element of the next generation of search engines, systems with the capacity to find "the need behind the query". This new generation will provide specialized services and interfaces according to the search domain and type of information needed. Moreover, it will integrate textual search (websites) and multimedia search (images, audio, video), it will be able to find and organize information, rather than generating ranked lists of websites.

Removing noisy mentions for distant supervision

Los métodos para Extracción de Información basados en la Supervisión a Distancia se basan en usar tuplas correctas para adquirir menciones de esas tuplas, y así entrenar un sistema tradicional de extracción de información supervisado. En este artículo analizamos las fuentes de ruido en las menciones, y exploramos métodos sencillos para filtrar menciones ruidosas. Los resultados demuestran que combinando el filtrado de tuplas por frecuencia, la información mutua y la eliminación de menciones lejos de los centroides de sus respectivas etiquetas mejora los resultados de dos modelos de extracción de información significativamente.

Towards a new proposal to evaluate the learning objects quality in learning strategies for education (QEES)

In Computer Science world several proposals have been developed for the assessment of the quality of the digital objects, based on the capabilities and facilities offered by current technologies and the available resources. Years ago researchers and specialists from both educational and technological areas have been committed to the development of strategies that improve the quality of education. At present, in the field of teaching-learning, another important aspect is the need to improve the manner of gaining knowledge and learning in education, which the use of learning strategies is a major advance in the teaching-learning process in institutions of higher education. This paper presents QEES, a proposal for evaluating the quality of the learning objects employed on learning strategies to support students during their education processes by using information extraction techniques and ontologies.

Enrichment of the Phenotypic and Genotypic Data Warehouse analysis using Question Answering systems to facilitate the decision making process in cereal breeding programs

Currently there are an overwhelming number of scientific publications in Life Sciences, especially in Genetics and Biotechnology. This huge amount of information is structured in corporate Data Warehouses (DW) or in Biological Databases (e.g. UniProt, RCSB Protein Data Bank, CEREALAB or GenBank), whose main drawback is its cost of updating that makes it obsolete easily. However, these Databases are the main tool for enterprises when they want to update their internal information, for example when a plant breeder enterprise needs to enrich its genetic information (internal structured Database) with recently discovered genes related to specific phenotypic traits (external unstructured data) in order to choose the desired parentals for breeding programs. In this paper, we propose to complement the internal information with external data from the Web using Question Answering (QA) techniques. We go a step further by providing a complete framework for integrating unstructured and structured information by combining traditional Databases and DW architectures with QA systems. The great advantage of our framework is that decision makers can compare instantaneously internal data with external data from competitors, thereby allowing taking quick strategic decisions based on richer data.

ADRSpanishTool: una herramienta para la detección de efectos adversos e indicaciones

Presentamos una herramienta basada en coocurrencias de fármaco-efecto para la detección de reacciones adversas e indicaciones en comentarios de usuarios procedentes de un foro médico en español. Además, se describe la construcción automática de la primera base de datos en español sobre indicaciones y efectos adversos de fármacos.

Using IR-n for information retrieval of Genomics Track

Nowadays there is a big amount of biomedical literature which uses complex nouns and acronyms of biological entities thus complicating the task of retrieval specific information. The Genomics Track works for this goal and this paper describes the approach we used to take part of this track of TREC 2007. As this is the first time we participate in this track, we configurated a new system consisting of the following diferenciated parts: preprocessing, passage generation, document retrieval and passage (with the answer) extraction. We want to call special attention to the textual retrieval system used, which was developed by the University of Alicante. Adapting the resources for the propouse, our system has obtained precision results over the mean and median average of the 66 official runs for the Document, Aspect and Passage2 MAP; and in the case of Passage MAP we get nearly the median and mean value. We want to emphasize we have obtained these results without incorporating specific information about the domain of the track. For the future, we would like to further develop our system in this direction.

Open business intelligence: on the importance of data quality awareness in user-friendly data mining

Citizens demand more and more data for making decisions in their daily life. Therefore, mechanisms that allow citizens to understand and analyze linked open data (LOD) in a user-friendly manner are highly required. To this aim, the concept of Open Business Intelligence (OpenBI) is introduced in this position paper. OpenBI facilitates non-expert users to (i) analyze and visualize LOD, thus generating actionable information by means of reporting, OLAP analysis, dashboards or data mining; and to (ii) share the new acquired information as LOD to be reused by anyone. One of the most challenging issues of OpenBI is related to data mining, since non-experts (as citizens) need guidance during preprocessing and application of mining algorithms due to the complexity of the mining process and the low quality of the data sources. This is even worst when dealing with LOD, not only because of the different kind of links among data, but also because of its high dimensionality. As a consequence, in this position paper we advocate that data mining for OpenBI requires data quality-aware mechanisms for guiding non-expert users in obtaining and sharing the most reliable knowledge from the available LOD.

Aplicaciones de procesamiento de lenguaje natural

El campo de procesamiento de lenguaje natural (PLN), ha tenido un gran crecimiento en los últimos años; sus áreas de investigación incluyen: recuperación y extracción de información, minería de datos, traducción automática, sistemas de búsquedas de respuestas, generación de resúmenes automáticos, análisis de sentimientos, entre otras. En este artículo se presentan conceptos y algunas herramientas con el fin de contribuir al entendimiento del procesamiento de texto con técnicas de PLN, con el propósito de extraer información relevante que pueda ser usada en un gran rango de aplicaciones. Se pueden desarrollar clasificadores automáticos que permitan categorizar documentos y recomendar etiquetas; estos clasificadores deben ser independientes de la plataforma, fácilmente personalizables para poder ser integrados en diferentes proyectos y que sean capaces de aprender a partir de ejemplos. En el presente artículo se introducen estos algoritmos de clasificación, se analizan algunas herramientas de código abierto disponibles actualmente para llevar a cabo estas tareas y se comparan diversas implementaciones utilizando la métrica F en la evaluación de los clasificadores.

The young open cluster Berkeley 55

We present UBV photometry of the highly reddened and poorly studied open cluster Berkeley 55, revealing an important population of B-type stars and several evolved stars of high luminosity. Intermediate-resolution far-red spectra of several candidate members confirm the presence of one F-type supergiant and six late supergiants or bright giants. The brightest blue stars are mid-B giants. Spectroscopic and photometric analyses indicate an age 50 ± 10 Myr. The cluster is located at a distance d ≈ 4 kpc, consistent with other tracers of the Perseus Arm in this direction. Berkeley 55 is thus a moderately young open cluster with a sizable population of candidate red (super)giant members, which can provide valuable information about the evolution of intermediate-mass stars.

3D Feature Extraction and Modelling for SLAM

Paper submitted to the 39th International Symposium on Robotics ISR 2008, Seoul, South Korea, October 15-17, 2008.

Green education using open educational resources (OER): setting up a green OER repository

A wealth of open educational resources (OER) focused on green topics is currently available through a variety of sources, including learning portals, digital repositories and web sites. However, in most cases these resources are not easily accessible and retrievable, while additional issues further complicate this issue. This paper presents an overview of a number of portals hosting OER, as well as a number of “green” thematic portals that provide access to green OER. It also discusses the case of a new collection that aims to support and populate existing green collections and learning portals respectively, providing information on aspects such as quality assurance/collection and curation policies, workflow and tools for both the content and metadata records that apply to the collection. Two case studies of the integration of this new collection to existing learning portals are also presented.

Real-time 3D semi-local surface patch extraction using GPGPU

Feature vectors can be anything from simple surface normals to more complex feature descriptors. Feature extraction is important to solve various computer vision problems: e.g. registration, object recognition and scene understanding. Most of these techniques cannot be computed online due to their complexity and the context where they are applied. Therefore, computing these features in real-time for many points in the scene is impossible. In this work, a hardware-based implementation of 3D feature extraction and 3D object recognition is proposed to accelerate these methods and therefore the entire pipeline of RGBD based computer vision systems where such features are typically used. The use of a GPU as a general purpose processor can achieve considerable speed-ups compared with a CPU implementation. In this work, advantageous results are obtained using the GPU to accelerate the computation of a 3D descriptor based on the calculation of 3D semi-local surface patches of partial views. This allows descriptor computation at several points of a scene in real-time. Benefits of the accelerated descriptor have been demonstrated in object recognition tasks. Source code will be made publicly available as contribution to the Open Source Point Cloud Library.

Shape Effect of Electrochemical Chloride Extraction in Structural Reinforced Concrete Elements Using a New Cement-Based Anodic System

This article shows the research carried out by the authors focused on how the shape of structural reinforced concrete elements treated with electrochemical chloride extraction can affect the efficiency of this process. Assuming the current use of different anode systems, the present study considers the comparison of results between conventional anodes based on Ti-RuO2 wire mesh and a cement-based anodic system such as a paste of graphite-cement. Reinforced concrete elements of a meter length were molded to serve as laboratory specimens, to closely represent authentic structural supports, with circular and rectangular sections. Results confirm almost equal performances for both types of anode systems when electrochemical chloride extraction is applied to isotropic structural elements. In the case of anisotropic ones, such as rectangular sections with no uniformly distributed rebar, differences in electrical flow density were detected during the treatment. Those differences were more extreme for Ti-RuO2 mesh anode system. This particular shape effect is evidenced by obtaining the efficiencies of electrochemical chloride extraction in different points of specimens.

New Approaches for Teaching Soil and Rock Mechanics Using Information and Communication Technologies

Soil and rock mechanics are disciplines with a strong conceptual and methodological basis. Initially, when engineering students study these subjects, they have to understand new theoretical phenomena, which are explained through mathematical and/or physical laws (e.g. consolidation process, water flow through a porous media). In addition to the study of these phenomena, students have to learn how to carry out estimations of soil and rock parameters in laboratories according to standard tests. Nowadays, information and communication technologies (ICTs) provide a unique opportunity to improve the learning process of students studying the aforementioned subjects. In this paper, we describe our experience of the incorporation of ICTs into the classical teaching-learning process of soil and rock mechanics and explain in detail how we have successfully developed various initiatives which, in summary, are: (a) implementation of an online social networking and microblogging service (using Twitter) for gradually sending key concepts to students throughout the semester (gradual learning); (b) detailed online virtual laboratory tests for a delocalized development of lab practices (self-learning); (c) integration of different complementary learning resources (e.g. videos, free software, technical regulations, etc.) using an open webpage. The complementary use to the classical teaching-learning process of these ICT resources has been highly satisfactory for students, who have positively evaluated this new approach.