Title :
An Axiomatic Approach to Constructing Distances for Rank Comparison and Aggregation
Author :
Farnoud, Farzad ; Milenkovic, Olgica
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Illinois at Urbana-Champaign, Urbana, IL, USA
Abstract :
We propose a new family of distance measures on rankings, derived through an axiomatic approach, that consider the nonuniform relevance of the top and bottom of ordered lists and similarities between candidates. The proposed distance functions include specialized weighted versions of the Kendall τ distance and the Cayley distance, and are suitable for comparing rankings in a number of applications, including information retrieval and rank aggregation. In addition to proposing the distance measures and providing the theoretical underpinnings for their applications, we also analyze algorithmic and computational aspects of weighted distance-based rank aggregation. We present an aggregation method based on approximating weighted distance measures by a generalized version of Spearman´s footrule distance as well as a Markov chain method inspired by PageRank, where transition probabilities of the Markov chain reflect the chosen weighted distances.
Keywords :
data reduction; information retrieval; Cayley distance; Kendall tau distance; Markov chain method; distance construction; distance functions; information retrieval; rank comparison; weighted distance based rank aggregation; Aggregates; Bioinformatics; Cities and towns; Measurement; Search engines; Tin; Transforms; PageRank; Weighted Kendall distance; collaborative filtering; information retrieval; positional relevance; rank aggregation; similarity; statistics; top-vs-bottom;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2014.2345760