122
reads

Shinto Eguchi & Osamu Komori  

2010-04-25 
11:00 - 12:00 

308 , Mathematics Research Center Building (ori. New Math. Bldg.)

Biotechnology enables us to observe simultaneously a huge dimensional data in such as microarray, mass spectrometry, SNIP typing and so forth. We apply several boosting algorithms to pattern recognition for a phenotype based on such high-dimensional feature vector. For example, if the phenotype is binary in expressing a treatment effect for patients, efficient prediction for the phenotype enables realization of personalized medicine. Several applications for real and synthetics data are given to confirm the statistical performance. Future problems are explored.

For material related to this talk, click here.