El grup de processament del llenguatge natural inicia la seva activitat al 1988. Des dels seus orígens ha sigut un grup interdisciplinari (informàtica i llingüistica). Ha portat a terme una prolífica activitat en diverses àrees d'investigació dins de processament del llenguatge natural i la intel•ligència artificial. En el processament bàsic de la llengua destaquen els camps de desambiguació morfosintàctica, anàlisis sintàctica i semàntica, desambiguació semàntica, etc., amb especial èmfasi en l'aplicació de mètodes estadístics i d'aprenentatge automàtic per resoldre aquestes tasques. Així mateix, apart de la tecnologia bàsica, també s'aborden aplicacions de més alt nivell, com ara traducció automàtica, extracció d'informació, donar resposta a preguntes, resum automàtic, processament de diccionaris, de corpus textuals, i de recursos lingüístics en general.

http://futur.upc.edu/GPLN

The Natural Language Processing Group (GPLN) conducts research into the automatic processing of natural language, particularly those aspects related to the construction and exploitation of multilingual lexical resources and the extraction of information from documents written in natural language and summarisation. Within these research lines, the GPLN focuses its attention on developing language processing tools, such as morphological and syntactical analysers, grammatical labelers, corpus processing systems, and specific machine learning algorithms for natural language processing and on compiling linguistic resources, such as grammars, dictionaries and lexical and conceptual databases. The GPLN has also been working on natural language software specifications for the development of tools for the quality control of software. The GPLN collaborates with the Speech Processing Group in the research of oral translation and the development of dialogue systems.

http://futur.upc.edu/GPLN

The Natural Language Processing Group (GPLN) conducts research into the automatic processing of natural language, particularly those aspects related to the construction and exploitation of multilingual lexical resources and the extraction of information from documents written in natural language and summarisation. Within these research lines, the GPLN focuses its attention on developing language processing tools, such as morphological and syntactical analysers, grammatical labelers, corpus processing systems, and specific machine learning algorithms for natural language processing and on compiling linguistic resources, such as grammars, dictionaries and lexical and conceptual databases. The GPLN has also been working on natural language software specifications for the development of tools for the quality control of software. The GPLN collaborates with the Speech Processing Group in the research of oral translation and the development of dialogue systems.

http://futur.upc.edu/GPLN

Enviaments recents

  • Boosting trees for anti-spam email filtering (Extended version) 

    Carreras Pérez, Xavier; Màrquez Villodre, Lluís (2001-10)
    Report de recerca
    Accés obert
    In this work, a set of comparative experiments for the problem of automatically filtering unwanted electronic mail messages are performed on two public corpora: PU1 and LingSpam. Several variants of the AdaBoost algorithm ...
  • Strategies to tackle problematic input in a Spanish dialogue system 

    Arranz Corzana, María Victoria; Castell Ariño, Núria; Civit Torruella, Montserrat (2001-11)
    Report de recerca
    Accés obert
    This paper focuses on the strategies adopted to tackle problematic input and ease communication between modules in a Spanish railway information dialogue system for spontaneous speech. The paper describes the design and ...
  • Analyzing temporal integrity constraints to obtain the minimum number of transition rules 

    Martín Escofet, Carme (2001-12)
    Report de recerca
    Accés obert
    The aim of this paper is to obtain advantages for integrity constraint checking in bitemporal deductive databases, analyzing the temporal integrity constraint behaviour. First at all, a new taxonomy of temporal integrity ...
  • Restriccions d'integritat temporals en bases de dades deductives bitemporals 

    Martín Escofet, Carme (2000-12)
    Report de recerca
    Accés obert
    The aim of this report is to introduce a taxonomy of temporal integrity constraints, focused on the bitemporal deductive database area, to get a better understanding of why they are required, their behavior and the best ...
  • A Proposal for wide-coverage Spanish named entity recognition 

    Arévalo, M.; Carreras Pérez, Xavier; Màrquez Villodre, Lluís; Martí Antonin, Maria Antònia; Padró, Lluís; Simon, Maria José (2002-04)
    Report de recerca
    Accés obert
    This paper presents a proposal for wide--coverage Named Entity Recognition for Spanish. First, a linguistic description of the typology of Named Entities is proposed. Following this definition an architecture of sequential ...
  • The Synonym management process in SAREL 

    Hernández Gómez, M. Angeles; Castell Ariño, Núria (2002-05)
    Report de recerca
    Accés obert
    The specification phase is one of the most important and least supported parts of the software development process. The SAREL system has been conceived as a knowledge-based tool to improve the specification phase. The ...
  • The Data warehouse: a temporal database 

    Martín Escofet, Carme; Abelló Gamazo, Alberto (2002-11)
    Report de recerca
    Accés obert
    The aim of this paper is to bring together two research areas, i.e. Data Warehouses and Temporal Databases, involving representation of time. In order to achieve this goal, data warehouse and temporal database research ...
  • Searching by approximate personal-name matching 

    Daude Ventura, Jordi; Camps Pare, Rafael (2003-04)
    Report de recerca
    Accés obert
    We discuss the design, building and evaluation of a method to access theinformation of a person, using his name as a search key, even if it has deformations. We present a similarity function, the DEA function, based on ...
  • Margin maximization with feed-forward neural networks: a comparative study with support vector machines and AdaBoost 

    Romero Merino, Enrique; Màrquez Villodre, Lluís; Carreras Pérez, Xavier (2003-06)
    Report de recerca
    Accés obert
    Feed-forward Neural Networks (FNN) and Support Vector Machines (SVM) are two machine learning frameworks developed from very different starting points of view. In this work a new learning model for FNN is proposed such ...
  • The Modifiability factor in the LESD project: definition and practical results 

    Castell Ariño, Núria; Slavkova Hernández, Ólga (1995-02)
    Report de recerca
    Accés obert
    The LESD project (Linguistic Engineering for Software Development) aims to develop computing tools for analysis and reasoning on functional or preliminary specifications of aerospace software written in English. These tools ...

Mostra'n més