Hello, Saw the announcement of Mahout and thought there might be the possibility of a symbiotic relationship here. I develop the recommendations system for StyleFeeder. Did my PhD thesis at MIT on collaborative filtering and have been working with/developing machine learning algorithms for over a decade now (geez, I'm getting old!). We may need to go parallel w/ the recs system here at StyleFeeder in the not-so-distant future and we're java-based and open source junkies, so if we can help Mahout while achieving that goal, all the better.
I'm experienced in the math behind ML algorithms and serial implementations thereof. Minimal parallel experience. Jason P.S. Cool that Mahout is based on an Andrew Ng paper. I have a tremendous amount of respect for him. Very sharp. -- Jason Rennie Head of Machine Learning Technologies, StyleFeeder http://www.stylefeeder.com/ http://people.csail.mit.edu/jrennie/
