Session Title
Map Reduce
Abstract
This tutorial will introduce the main concepts of the Map-Reduce paradigm and the corresponding underlying problem-solving strategies. We will start by explaining the basic concepts of the Map-Reduce framework, namely mappers, reducers, external sorting and grouping. Then, we will show how the Map-Reduce framework allows easy scaling of data-intensive computation tasks. Finally, we will show how to "Map-Reduce" several prototypical problems and applications ranging from simple dictionary and index creation in large document collections, to the computation of statistics on large data sets, and graph-manipulation algorithms.
Language
Portuguese
Speakers
Luís Sarmento
PhD student in Software Engineering at FEUP under FCT scholarship SFRH/BD/23590/2005, and I am a member of the NIAD&R team. I
His research is related to studying and developing methods for measuring semantic similarity (or dissimilarity) between words (names and verbs) or concepts (identified by noun phases or names). The goal of this research is being able to use such semantic information for helping to build more powerful semi-supervised methods for NLP/IE tasks. (more)
Where
When
Friday, 4 of December of 2009, from 10:00 to 11:00
Files and video