Twitter LinkedIn
Return to Wellspring

Projects Sharing Researchers

Communicate

Computational Gaming: A technique for improving the collection of unbiased data labels

Details

Project TitleComputational Gaming: A technique for improving the collection of unbiased data labels
Track Code2010-081
Short DescriptionNone
AbstractNone
 
TagsNone
 
Posted DateOct 19, 2011 4:24 PM

Manager

Name
Gorana Smailagic

What is Computational Gaming?

Computational Gaming is a technique for improving the collection of unbiased data labels.

Computational Gaming builds on machine learning techniques by using gaming to collect data from users, rather than using paid forms of crowdsourcing, for example, Mechanical Turk. Paying users naturally biases responses from users. The novel contributions of Computational Gaming allow the more efficient and more effective collection of labels for datasets, that can be applied to a large number of domains.

Contact Information

Please contact Gorana Smailagic, gorana@cmu.edu, if you'd like to know more about this technology.