Skip to main content

ADQUISICION DE ESCENARIOS DE CONOCIMIENTO A TRAVES DE LA LECTURA DE TEXTOS: INFERENCIA DE RELACIONES ENTRE EVENTOS

TIN2012-38584-C06-01

•
Funding agency name Ministerio de Economía y Competitividad
Funding agency acronym MINECO
Program Programa Nacional de Proyectos de Investigación Fundamental
Subprogram Subprograma de Proyectos de Investigación Fundamental no Orientada
Call Proyectos I+D
Call year 2012
State Plan 2008-2011 Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica
Management unit Secretaría de Estado de Investigación, Desarrollo e Innovación
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 9
Encontrada(s) 1 página(s)

Repositorio Digital de la UPF
Publication Artículo científico (JournalArticle)

Robust multilingual Named Entity Recognition with shallow semi-supervised features

  • Agerri, Rodrigo
  • Rigau Claramunt, German
We present a multilingual Named Entity Recognition approach based on a robust and general set of features across languages and datasets. Our system combines shallow local information with clustering semi-supervised features induced on large amounts of unlabeled text. Understanding via empirical experimentation how to effectively combine various types of clustering features allows us to seamlessly export our system to other datasets and languages. The result is a simple but highly competitive system which obtains state of the art results across five languages and twelve datasets. The results are reported on standard shared task evaluation data such as CoNLL for English, Spanish and Dutch. Furthermore, and despite the lack of linguistically motivated features, we also report best results for languages such as Basque and German. In addition, we demonstrate that our method also obtains very competitive results even when the amount of supervised data is cut by half, alleviating the dependency on manually annotated data. Finally, the results show that our emphasis on clustering features is crucial to develop robust out-of-domain models. The system and models are freely available to facilitate its use and guarantee the reproducibility of results., This work has been supported by the European projects NewsReader, EC/FP7/316404 and QTLeap – EC/FP7/610516, and by the Spanish Ministry of Economy and Competitiveness (MINECO) SKATER, Grant No. TIN2012-38584-C06-01 and TUNER, TIN2015-65308-C5-1-R.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Methodology and evaluation of the Galician WordNet expansion with the WN-Toolkit, Metodología y evaluación de la expansión del WordNet del gallego con WN-Toolkit

  • Gómez Guinovart, Xavier
  • Oliver González, Antoni
In this paper the methodology and a detailed evaluation of the results of the expansion of the Galician WordNet using the WN-Toolkit are presented. This toolkit allows the creation and expansion of wordnets using the expand model. In our experiments we have used methodologies based on dictionaries and parallel corpora. The evaluation of the results has been performed both in an automatic and in a manual way, allowing a comparison of the precision values obtained with both evaluation procedures. The manual evaluation provides details about the source of the errors. This information has been very useful for the improvement of the toolkit and for the correction of some errors in the reference WordNet for Galician.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

ParTes. Test Suite for Parsing Evaluation, ParTes: Test suite para evaluación de analizadores sintácticos

  • Lloberes Salvatella, Marina
  • Castellón Masalles, Irene
  • Padró Cirera, Lluís
  • Gonzàlez Pellicer, Edgar
This paper presents ParTes, the first test suite in Spanish and Catalan for parsing qualitative evaluation. This resource is a hierarchical test suite of the representative syntactic structure and argument order phenomena. ParTes proposes a simplification of the qualitative evaluation by contributing to the automatization of this task.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Boosting Terminology Extraction through Crosslingual Resources, Mejora de la extracción de terminología usando recursos translingües

  • Cajal, Sergio
  • Rodríguez Hontoria, Horacio
Terminology Extraction is an important Natural Language Processing task with multiple applications in many areas. The task has been approached from different points of view using different techniques. Language and domain independent systems have been proposed as well. Our contribution in this paper focuses on the improvements on Terminology Extraction using crosslingual resources and specifically the Wikipedia and on the use of a variant of PageRank for scoring the candidate terms.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Adaptive generation of multilingual questions and answers from web content

  • Gatius Vila, Marta|||0000-0003-0456-7912
(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works., This article describes the adaptive generation of questions and answers from web content to assist the user when searching for information. The approach proposed integrates user models, domain-restricted knowledge representation, natural language and wrappers techniques. This proposal is based on the representation of the main concepts involved in a specific scenario as a set of attributes, and the user models as the subset of those conceptual attributes relevant for each user. The approach described has been applied to generate multilingual personalized questions and answers from web sites in two different domains: university courses and dissemination on nutrition.
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Unsupervised ensemble minority clustering

  • Gonzàlez Pellicer, Edgar
  • Turmo Borras, Jorge|||0000-0002-7521-1115
Cluster analysis lies at the core of most unsupervised learning tasks. However, the majority of clustering algorithms depend on the all-in assumption, in which all objects belong to some cluster, and perform poorly on minority clustering tasks, in which a small fraction of signal data stands against a majority of noise. The approaches proposed so far for minority clustering are supervised: they require the number and distribution of the foreground and background clusters. In supervised learning and all-in clustering, combination methods have been successfully applied to obtain distribution-free learners, even from the output of weak individual algorithms. In this work, we propose a novel ensemble minority clustering algorithm, EWOCS, suitable for weak clustering combination. Its properties have been theoretically proved under a loose set of constraints. We also propose a number of weak clustering algorithms, and an unsupervised procedure to determine the scaling parameters for Gaussian kernels used within the task. We have implemented a number of approaches built from the proposed components, and evaluated them on a collection of datasets., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Unsupervised ensemble minority clustering

  • Gonzàlez Pellicer, Edgar
  • Turmo Borras, Jorge
Cluster analysis lies at the core of most unsupervised learning tasks. However, the majority of clustering algorithms depend on the all-in assumption, in which all objects belong to some cluster, and perform poorly on minority clustering tasks, in which a small fraction of signal data stands against a majority of noise. The approaches proposed so far for minority clustering are supervised: they require the number and distribution of the foreground and background clusters. In supervised learning and all-in clustering, combination methods have been successfully applied to obtain distribution-free learners, even from the output of weak individual algorithms. In this work, we propose a novel ensemble minority clustering algorithm, EWOCS, suitable for weak clustering combination. Its properties have been theoretically proved under a loose set of constraints. We also propose a number of weak clustering algorithms, and an unsupervised procedure to determine the scaling parameters for Gaussian kernels used within the task. We have implemented a number of approaches built from the proposed components, and evaluated them on a collection of datasets., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Adaptive generation of multilingual questions and answers from web content

  • Gatius Vila, Marta
(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works., This article describes the adaptive generation of questions and answers from web content to assist the user when searching for information. The approach proposed integrates user models, domain-restricted knowledge representation, natural language and wrappers techniques. This proposal is based on the representation of the main concepts involved in a specific scenario as a set of attributes, and the user models as the subset of those conceptual attributes relevant for each user. The approach described has been applied to generate multilingual personalized questions and answers from web sites in two different domains: university courses and dissemination on nutrition., Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Robust multilingual Named Entity Recognition with shallow semi-supervised features

  • Agerri, Rodrigo
  • Rigau Claramunt, German
We present a multilingual Named Entity Recognition approach based on a robust and general set of features across languages and datasets. Our system combines shallow local information with clustering semi-supervised features induced on large amounts of unlabeled text. Understanding via empirical experimentation how to effectively combine various types of clustering features allows us to seamlessly export our system to other datasets and languages. The result is a simple but highly competitive system which obtains state of the art results across five languages and twelve datasets. The results are reported on standard shared task evaluation data such as CoNLL for English, Spanish and Dutch. Furthermore, and despite the lack of linguistically motivated features, we also report best results for languages such as Basque and German. In addition, we demonstrate that our method also obtains very competitive results even when the amount of supervised data is cut by half, alleviating the dependency on manually annotated data. Finally, the results show that our emphasis on clustering features is crucial to develop robust out-of-domain models. The system and models are freely available to facilitate its use and guarantee the reproducibility of results., This work has been supported by the European projects NewsReader, EC/FP7/316404 and QTLeap – EC/FP7/610516, and by the Spanish Ministry of Economy and Competitiveness (MINECO) SKATER, Grant No. TIN2012-38584-C06-01 and TUNER, TIN2015-65308-C5-1-R.