Skip to main content
. Author manuscript; available in PMC: 2016 Apr 19.
Published in final edited form as: Pac Symp Biocomput. 2016;21:144–155.

FIGURE 1.

FIGURE 1

A data matrix is approximated as the product of two matrices. By construction, the resulting approximation is of lower algebraic rank. The data matrix A may contain features of different data-types and missing entries, as illustrated here. Each row of X is an encoding of an observation in A in the latent feature space. Each column of Y is an encoding of a feature of A in the latent feature space.