Hi, I'm working with a really sparse matrix (way less than 1% sparse) trying to infer missing values with Matrix Factorization. Is there any theory on "accuracy" with really sparse data or any other techniques that purport to perform on this type of matrix? thanks in advance, Jonny

asked Jul 30 '12 at 16:05

Jonny%20Edwards's gravatar image

Jonny Edwards
1111

I never dealt with the problem myself but may be you want to check Emmanuel Candes and Yi Ma's work on low rank matrix factorization.

http://www-stat.stanford.edu/~candes/papers/MatrixOracle.pdf

http://perception.csl.illinois.edu/matrix-rank/Files/RPCA_JACM.pdf

(Jul 30 '12 at 17:00) Rakesh

I would imagine that all the work for the NetFlix competition (inferring the missing values of a user, value => score matrix) are relevant here.

(Aug 05 '12 at 05:19) Joseph Turian ♦♦
Be the first one to answer this question!
toggle preview

powered by OSQA

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.