====== Intelligent Extraction and Retrieval of lexico-semantic patterns in oncology texts in English ======

FUNDING ENTITY: Fundación Séneca

BUDGET: 625000 ptas

LENGTH: 2002-2004

PARTNERS:

Universidad de Murcia

PRIME INVESTIGATOR:

Pascual Cantos Gómez

OTHER RESEARCHERS:

Pascual Pérez Paredes
Fernando Martín Rubio
Rodrigo Martínez Béjar
Juan García Iborra
Laura María Campoy Gómez
Jesualdo Tomás Fernández Breis
Manuel De las Heras González
Juan Salinas Ramos
Isabel De la Fuente Muñoz

ABSTRACT

This project aims at facilitating the selective access to information by using the most advanced technologies in (1) computational lexicography (lexical constellations) and (2) knowledge arquitectures to serve other disciplines such as oncology. The model of lexical constellations allows for the identification of collocational data, and determining the higher and more complex lexico-semantic structures: thesaurus. The reliability and usefulness of the model and its computability allow for the automatic extraction of valid and relevant collocational data and lexico-semantic patterns.

 

OBJECTIVES

To collect oncological texts in English to compile and exhaustive and representative linguistic corpus in this domain.

To obtain the most efficient method to identify collocational data (lexico-semantic patterns) in medical-oncological English.

Validation of the initial collocational data to classify them according their different lexico-semantic values, and detection of composed expresions and/or oncological idioms.

Determining the influence of behaviour patterns and the socialization of words to proceed to represent ontologically terms and idioms.

Implementation of the clasification and the ontology of terms and idioms in a relational database.

Obtention of a model of an organizational memory of the oncological English.

Implementation of a system to extract oncological knowledge from the web, which must be valid, reliable and highly operative; it must be orientes to professionals and researchers in oncology.