On Ordinal Covering of Proposals Using Balanced Incomplete Block Designs

Computer Science – Discrete Mathematics

Scientific paper

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

Scientific paper

A frequently encountered problem in peer review systems is to facilitate pairwise comparisons of a given set of proposals by as few as referees as possible. In [8], it was shown that, if each referee is assigned to review k proposals then ceil{n(n-1)/k(k-1)} referees are necessary and ceil{n(2n-k)/k^2} referees are sufficient to cover all n(n-1)/2 pairs of n proposals. While the upper bound remains within a factor of 2 of the lower bound, it becomes relatively large for small values of k and the ratio of the upper bound to the lower bound is not less than 3/2 when 2 <= k <= n/2. In this paper, we show that, if sqrt(n) <= k <= n/2 then the upper and lower bounds can be made closer in that their ratio never exceeds 3/2. This is accomplished by a new method that assigns proposals to referees using a particular family of balanced incomplete block designs. Specifically, the new method uses ceil{n(n+k)/k^2} referees when n/k is a prime power, n divides k^2, and sqrt(n) <= k <= n/2. Comparing this new upper bound to the one given in [8] shows that the new upper bound approaches the lower bound as k tends to sqrt(n) whereas the upper bound in [8] approaches the lower bound as k tends to n. Therefore, the new method given here when combined together with the one in [8] provides an assignment whose upper bound referee complexity always remains within a factor of 3/2 of the lower bound when sqrt(n) <= k <= n, thereby improving upon the assignment described in [8]. Furthermore, the new method provides a minimal covering, i.e., it uses the minimum number of referees possible when k = sqrt(n) and k is a prime power.

No associations

LandOfFree

Say what you really think

Search LandOfFree.com for scientists and scientific papers. Rate them and share your experience with other people.

Rating

On Ordinal Covering of Proposals Using Balanced Incomplete Block Designs does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with On Ordinal Covering of Proposals Using Balanced Incomplete Block Designs, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and On Ordinal Covering of Proposals Using Balanced Incomplete Block Designs will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFWR-SCP-O-485567

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.