Hadoop Machine Learning Project

Hadoop Machine Learning Project

Project sponsored by Austin ACM SIGKDD in conjunction with the Hadoop Based Machine Learning Course to get hands on experience with Hadoop and machine learning.

The project will explore multiple methods to implement least squares regression with Hadoop:

1. Least squares solved with SVD, stochastic method in Mahout.

2. Least squares solved with QR factorization.

3. Least squares solved with stochastic sub-gradient method.

4. Least squares solved with bag of little bootstrap.

The methods will be related to discussions from the class. If we get all five methods implemented by May, I have a couple of additional methods we can explore. We can also look at other regression techniques such as logistic regression or generalized linear models.

Magia Pizza has free parking, free wifi, and a big screen TV we can plug into. Mangia Pizza asks that we have at least 10 people at the meeting and order food for free use of their meeting room.