P
Initializing...
Iterative assortment planning estimates relative preferences correctly: $\hat v_i(x)/\hat v_j(x)=v_i/v_j$ · Prove2Me