| Email: | baden | @ | ucsd. |
edu |
Obtaining knowledge from the prodigious output of numerical simulations and scientific experiments is a challenging problem. The process is heavily application dependent, and existing tools for handling data do not scale. This course will explore cross cutting issues in this growing area, discuss fruitful venues for research, and carry out case studies of motivating applications.
We'll cover spatial query techniques and access methods, feature extraction and tracking, compression, and hardware acceleration (i.e. Cell BE and GPUs). The course will culminate in a research project.
In the past offering of this course, students developed techniques to analyze fluid flow simulations including database support and machine learning techniques, examined novel ways of compressing data, and explored facial recognition.
Students from outside CSE are welcome, including those with a mathematical or computational background.
This quarter we will have real world datasets available for study, including simulations of turbulent fluid flow, and digital microscopic data. Students may contribute their own data, including movie and photographic media.
| Maintained by | baden | @ | ucsd. |
edu | [ Tue Feb 10 20:27:27 PST 2009] |