Please use this identifier to cite or link to this item:
|Title:||Channel coding perspective of recommendation systems|
|Citation:||2009 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, VOLS 1- 4,319-323|
|Abstract:||Motivated by recommendation systems, we consider the problem of estimating block constant binary matrices (of size m x n) from sparse and noisy observations. The observations are obtained from the underlying block constant matrix after unknown row and column permutations, erasures, and errors. We derive upper and lower bounds on the achievable probability of error. For fixed erasure and error probability, we show that there exists a constant C(1) such that if the cluster sizes are less than C(1) In(mn), then for any algorithm the probability of error approaches one as m,n -> 77 infinity. On the other hand, we show that a simple polynomial time algorithm gives probability of error diminishing to zero provided the cluster sizes are greater than C(2) In(mn) for a suitable constant C(2).|
|Appears in Collections:||Proceedings papers|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.