Information : cette page n'est pas traduite en français
Sabine McConnell : Challenge and opportunities of the analysis of Petabyte-sized datasets for scientific applications.
Jeudi 22 novembre 2012, LIRMM
Sabine McConnell, Queen's University in Kingston, Canada
Title: Challenge and opportunities of the analysis of Petabyte-sized datasets for scientific applications.
Abstract: In the past two decades, data mining, often defined as the extraction of novel information from large datasets, has established itself as a valuable tool in a vast range of research areas. Businesses applications now routinely handle Big Data: heterogeneous datasets of Terabyte and Petabyte sizes. Techniques to facilitate the analysis and mining of such datasets include sampling, incremental techniques, as well as parallel and distributed implementations, with mixed results for typical scientific applications that still target much smaller datasets. I will provide an overview of existing techniques, with a focus on the challenges and opportunities as we move towards the analysis of Petabyte-sized datasets for scientific applications.
Bio: Sabine McConnell received her PhD in Computer Science from Queen's University in Kingston, Canada. Currently, She is an Associate Professor in the Department of Computing and Information Systems at Trent University in Peterborough, Canada. Her research focuses on the intersection of data mining, parallel implementations and astronomical applications. Most recently, She has worked on the the scalability of implementations of particular data mining techniques on clusters of accelerators.
Dernière mise à jour le 19/06/2013