Project suggestions for CS 4033/5033 Fall 2010

The following list of project suggestions is intended to get your project ideas started. Please feel free to come up with your own project ideas as well! Also, feel free to take one of these projects and modify it. Examine all of the previous years' suggestions for even more project ideas.

Atmospheric contaminants

• The AI committee at the annual AMS meeting has a contest every year on using the best AI/ML methods to predict an interesting weather task. This year's task is to predict the hourly levels of ozone and also to identify days with severe ozone conditions. Accurate prediction of these tasks are critical to many people with asthma and allergies or other breathing difficulties.
• The data is drawn from the real world: measurements from Rome, Italy during 2006.
• Often there is a monetary prize associated with winning this contest.

Educational data

• Why do some students persist in Science Technology Engineering and Mathematics majors while others do not? If you are interested in this question, we have an ongoing project in engineering education and a dataset of several thousand students in STEM majors. You could use ML methods to study this dataset.

Gaming

• Gaming is popular every year! Pick your favorite game where exhaustive search is NOT a feasible approach to solving the game. Look for an existing simulator (I don't want you spending time writing a simulator!) and then propose a way to incorporate learning into the game. Most often this takes the form of making an AI player.

Predicting Oklahoma weather

• The Oklahoma Mesonet has been observing the weather in all counties in Oklahoma for over 15 years. Temperature, air pressure, wind, and other observations are recorded every 5 minutes for each station. There are a variety of interesting tasks you could do with this dataset - talk to me if you are interested in using it.

Public health prediction

• Many public health datasets have properties that can be examined and predicted using machine learning.
  • One such data set is from the National Longitudinal Study of Adolescent Health.
  • Another data set is the Oklahoma Department of Public Health. I have a few of these data sets available (talk to me if you want to use them) but you can also download lots of aggregate statistics for free.
  • The Vaccine Adverse Event Reporting System is another public health data set that you can download.
  • There are tons more - if you have an interest in public health data, you can likely find a publically available data set to use for learning.

iPhone application

• We have a educational licene for iPhone development and anyone who wants to try iPhone app development using ML is welcome. Since this is a hard platform to write for unless you already know objective C, you'll need to talk to me about this but it could be neat!

This page written by amcgovern [at] ou.edu.

Last updated: May 23, 2016 1:59 PM