On Tue, Mar 31, 2009 at 11:43 PM, Atul Kulkarni <[email protected]> wrote: > questions in line. > > On Wed, Apr 1, 2009 at 1:27 AM, Ted Dunning <[email protected]> wrote: > >> Nobody is working on SVD yet, but one GSOC applicant has said that they >> would like to work on LDA which is a probabilistic relative of SVD. >> > I do not understand the relation in LDA and SVD. In my limited understanding > I understand LDA transforms data points in to a coordinate system where > they can be easily discriminated/classified. SVD on the other hand is used > for dimension reduction, can you help me bridge the gap by providing > something to read on?
LDA is an overloaded term. To the frequentist, it usually means Linear Discriminant Analysis, which is what you're talking about; to the bayesian machine learning people, it means Latent Dirichlet Allocation, which is a probabilistic dimensionality reduction technique for projecting documents in V-dimensional space to the K-simplex, with K \ll V. -- David > > >> The approach in your reference (3) is highly amenable to parallel >> implementation. > > Yes, I felt so too, but again did not want to comment on it untill I had the > MapReduce basics related with it. > >> >> >> Large-scale SVD would be a very interesting application for Mahout. >> > >> >> On Tue, Mar 31, 2009 at 11:09 PM, Atul Kulkarni <[email protected] >> >wrote: >> >> > Is there anyone doing the SVD part or are their any SVD algorithm >> > implementation on Hadoop? If there are then I would like to implement the >> > methods described in [1],[2],[3] for matrix factorization, in specific. >> > >> >> >> -- >> Ted Dunning, CTO >> DeepDyve >> > > > > -- > Regards, > Atul Kulkarni > Teaching Assistant, > Department of Computer Science, > University of Minnesota Duluth > Duluth. 55805. > www.d.umn.edu/~kulka053 >
