Advisor:   Anders Høst-Madsen | ahm@hawaii.edu

Prerequisites:  

Courses:  EE496

Focus:  

Description:  

We are working on methods of Data Discovery to extract interesting subsets of data out of big data sets. The application is for example to find subtle signs of disease from the large amount of data collected by smartwatches or fitness trackers. Or, to find automatic ways of making discoveries like this recent news: http://www.nytimes.com/2015/11/03/health/death-rates-rising-for-middle-aged-white-americans-study-finds.html?_r=0.

The task will be to develop computationally efficient implementations of the algorithms, preferably in Python (although we might instead use C++).

Requirements:

Students on the project should be good programmers, and have knowledge of some of the following, or be able to quickly learn:

- Able to write well-structured, modular, readable code with full documentation.

- Able to write very computationally efficient programs with things like memory management and hashing.

- Knowledge about parallel executable code.

- Knowledge of writing software in Python.

- Knows Matlab reasonably well. The Python programs should be callable from inside Matlab.

- Has some knowledge of probability theory and statistics (e.g., EE342).

Project Website:  http://ee.hawaii.edu/~madsen/Anders_Host-Madsen/Big_Data.html