I noticed that the estimation error correction in Algorithm 3 is not fairly specified. Since the base sets selection is not satisfied with 2^(r-1)<mi<=2^r in Kawamura's paper, the generality of your selection may need to be proved. And the notation bias(ck) is not specified either.
Edited 1 time(s). Last edit at 18-Sep-2011 14:23 by Orr.