RankDCG: Rank-Ordering Evaluation Measure

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Katerenchuk, Denys, Rosenberg, Andrew
Format: Preprint
Published: 2018
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911655419445248
author Katerenchuk, Denys
Rosenberg, Andrew
author_facet Katerenchuk, Denys
Rosenberg, Andrew
contents Ranking is used for a wide array of problems, most notably information retrieval (search). There are a number of popular approaches to the evaluation of ranking such as Kendall's $τ$, Average Precision, and nDCG. When dealing with problems such as user ranking or recommendation systems, all these measures suffer from various problems, including an inability to deal with elements of the same rank, inconsistent and ambiguous lower bound scores, and an inappropriate cost function. We propose a new measure, rankDCG, that addresses these problems. This is a modification of the popular nDCG algorithm. We provide a number of criteria for any effective ranking algorithm and show that only rankDCG satisfies all of them. Results are presented on constructed and real data sets. We release a publicly available rankDCG evaluation package.
format Preprint
id arxiv_https___arxiv_org_abs_1803_00719
institution arXiv
publishDate 2018
record_format arxiv
spellingShingle RankDCG: Rank-Ordering Evaluation Measure
Katerenchuk, Denys
Rosenberg, Andrew
Information Retrieval
Social and Information Networks
Ranking is used for a wide array of problems, most notably information retrieval (search). There are a number of popular approaches to the evaluation of ranking such as Kendall's $τ$, Average Precision, and nDCG. When dealing with problems such as user ranking or recommendation systems, all these measures suffer from various problems, including an inability to deal with elements of the same rank, inconsistent and ambiguous lower bound scores, and an inappropriate cost function. We propose a new measure, rankDCG, that addresses these problems. This is a modification of the popular nDCG algorithm. We provide a number of criteria for any effective ranking algorithm and show that only rankDCG satisfies all of them. Results are presented on constructed and real data sets. We release a publicly available rankDCG evaluation package.
title RankDCG: Rank-Ordering Evaluation Measure
topic Information Retrieval
Social and Information Networks
url https://arxiv.org/abs/1803.00719