You've built a vibrant community of Family Guy enthusiasts. The SVD recommendation algorithm took your site to the next level by allowing you to leverage the implicit knowledge of your community. But now you're ready for the next iteration - you are about
Stochastic matrices are easy to get -- just normalize the rows. Doubly stochastic matrices require more work -- simply normalizing columns/rows will not converge may take few dozen iterations to converge. One approach that works is to do constrained optim