Skip to main content
Ir a la página de inicio de la Comisión Europea (se abrirá en una nueva ventana)
español es
CORDIS - Resultados de investigaciones de la UE
CORDIS
Contenido archivado el 2024-05-29

A Distributed Data Mining Infrastruture

Objetivo

The collection and analysis of data play a major role in a broad range of economic, academic and scientific activities. Market research, economic policy, finance and engineering are increasingly reliant on large information resources, and the correct inter pretation of this data. Where the volume of data is exceedingly large, where it exists at disparate locations, or where the analysts are geographically dispersed, existing data mining techniques struggle to allow users to interpret their huge amounts of da ta. These problems exist to some extent for all research establishments and large businesses.We propose to develop a distributed data mining infrastructure (ADMIRE) which will provide more accurate decision-making tools and improve the quality of service ( QoS) for users. ADMIRE will be prototyped on two application types; from the scientific and commercial domains. These applications have been chosen for their intrinsic user service provision merit and their reliance on remote data access. The ADMIRE projec t aims to develop novel and innovative distributed data mining (DDM) techniques to deal with very large and distributed datasets. These datasets can then be mined to identify local and global trends and models. The ADMIRE project brings together separate s trands of computer science research in data mining, knowledge discovery, distributed and parallel computing, and application development tools. ADMIRE will be a self-configuring system that combines efficiency and heterogeneity of data. Efficiency is expre ssed in terms of quality of service, ease of use, short response time, and accuracy in decision-making. Heterogeneity is addressed at two levels: location of the datasets and format independence. ADMIRE requires a rethinking of how data mining techniques c an deal with large datasets, heterogeneity, and distributed locations of the datasets. This will result in new scalable algorithms and new models for data mining applications.

Ámbito científico (EuroSciVoc)

Palabras clave

Tema(s)

Convocatoria de propuestas

FP6-2002-MOBILITY-3
Consulte otros proyectos de esta convocatoria

Régimen de financiación

TOK - Marie Curie actions-Transfer of Knowledge

Coordinador

NATIONAL UNIVERSITY OF IRELAND, DUBLIN
Aportación de la UE
Sin datos
Coste total
Sin datos