Hello everybody, I know it's a little bit late, but I'm really excited to submit a google summer of code proposal for the Mahout project. I've read there is already a k-means implementation, so I've decided to implement another algorithm. I've chosen to implement one clustring and one classification algorithm, a priori the EM and SVM algorithms.
I've graduated college last december and I worked on multi-core parallel programing tools and implemented a parallel version of the k-means algorithm. I've just started my mastering course at the State University of Campinas (São Paulo, Brazil), mainly focused on content-based image retrieval and relevance feedback. So I'm really interested on clustering and classification algorithms since it's intensively used in image retrieval and an important optimization problem. Well, I have some java development skills but I've no experience with the Hadoop platform. What do you think about it? I do appreciate any comment and help. best, -- Rodrigo Tripodi skype id: rtripodi
