STATISTICS AND APPLIED PROBABILITY
SEMINAR
Place: NS Bldg, RM 234,
Time: Tuesday Mar 29, at 1:15-2:00
Speaker: John Schwarz, Math UofL
Title: "Overview of Clustering Algorithms with Applications to
MicroArray Data "
Abstract:
Statistical clustering can be applied to a number of large data
problems. The main idea behind the clustering technique is to
identify
values of the data with similar properties. There are a number of
different algorithms to identify clusters. Different algorithms can
alter the results by the method in which they are computed. Four
different categories of clustering will be examined that include;
hierarchical, k-means, mixed Gaussian and fuzzy clustering. The
purpose, methodology and brief synopses of the advantages and
disadvantages of each method will be discussed. Some examples of the
relevant software use in the analysis of the actual data
shall be
given.