Kernel functions and methods for analyzing rank data
Switch branches/tags
Nothing to show
Clone or download
Latest commit 6559295 Jun 26, 2018

README.md

kernrank

Travis-CI Build Status

kernrank is an R package implementing kernel functions and kernel methods for analyzing rank data, typically total rankings (or permutations), interleaving and top-k partial rankings, multivariate rankings.

Installation

# install.packages("devtools")
devtools::install_github("YunlongJiao/kernrank")

Demo

Please refer to GitHub repos kendallkernel and weightedkendall for a variety of examples of using the package.

References

Yunlong Jiao, Jean-Philippe Vert. "The Kendall and Mallows Kernels for Permutations." IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol. 40, no. 7, pp. 1755-1769, 2018. DOI:10.1109/TPAMI.2017.2719680

Yunlong Jiao, Jean-Philippe Vert. "The Weighted Kendall and High-order Kernels for Permutations." arXiv preprint arXiv:1802.08526, 2018. arXiv:1802.08526

Note

This package is built upon and is a stable and significant extension of the R CRAN package RMallow.

Misc