Numeerisen analyysin ja laskennallisen
tieteen seminaari
7.2.2005 klo
14.15
U322
Saara Hyvönen, Helsingin
Yliopisto, Tietojenkäsittelytieteen laitos
Data
Mining: Report from the Trenches
Data mining is application of mathematical, statistical and
computational tools
to analyze large data sets. Development in measurement and data
collection
technologies has made it possible to gather and store vast amounts of
data
in many areas of science and industry. However, our ability to find
information
from this data has increased at a much slower speed, and is therefore
naturally the
object of active research. We discuss data mining problems in theory
and practice by focusing on the analysis of two specific data
sets:
(1) atmospheric data (2) spatial data.
We introduce a number of methods applied to these data sets and results
obtainded,
as well as challenges posed by the analysis of real data sets.