clustering items and users by latent factors?

By SVD model, we can calculate latent factors for users and items. p(u) is latent factor for u, while q(i) is latent factor for i.

Recently, I thought about calculate user similarity by latent factor,

For example,

s(u,v) = f(p(u), p(v)) ?

I am testing this idea now, and I hope this idea can improve prediction accuracy.

PS.
I have tested this method on the residual of NSVD model. By using this clustering method in estimating group effects, I reduce the RMSE of NSVD model from 0.8923 to 0.8910