Compute the weighted Kendall distance (no ties) in O(n lg n)
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
main.cpp
readme.txt
weightedkendall.cpp
weightedkendall.h

readme.txt

Compute the weighted Kendall distance (no ties) in O(n lg n)
By F. Farnoud, July 2014

Gives the weighted Kendall distance between two rankings. 
Based on An Axiomatic Approach to Constructing Distances for Rank Comparison and Aggregation, F. Farnoud, and O. Milenkovic, IEEE Trans. Information Theory, Oct. 2014.

Use -h fol help.

NAME
        weightedkendall

DESCRIPTION
        Gives the weighted Kendall distance between two rankings.
        Based on An Axiomatic Approach to Constructing Distances
        for Rank Comparison and Aggregation, F. Farnoud,
        and O. Milenkovic, IEEE Trans. Information Theory, Oct. 2014.
USAGE 1:
        WeightedKendallDistance -f input-file
NOTES:
        With -f input, the first line is n (the length of the rankings), the second line is p1,
        the third line is p2, the fourth line is w.