Rank aggregation methods for the Web

By Cynthia Dwork, Ravi Kumar, Moni Naor, D. Sivakumar. 2001

2,068 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics, Information Retrieval & Search

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Statistical and Nonlinear Physics

In Information Retrieval & Search

Citations in the corpus

A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark (2019, methodology): “Therefore, we explore a few simple alternative ranking strategies (see Dwork et al. (2001) for an introduction raking aggregation methods): (i) Ranking according to the geometric mean.”
Ranking with Fairness Constraints (2017, background): “Alternate approaches mix together or aggregate diﬀerent rankings, e.g., as generated by diﬀerent interpretations of a query [RBCJ09, DKNS01].”
A knowledge plane for the internet (2003, background): “Examples include epidemic algorithms [5] for distributing data, Bayesian networks for learning [4], rank aggregation to enable a web of trust [6], constraint satisfaction algorithms [21], and policy-based management techniques [23,24].”
Optimal full ranking from pairwise comparisons (2021, background): “This highdimensional statistical estimation problem has important applications in many areas such as recommendation systems [1, 7], sports and gaming [2, 17, 23, 40, 41, 49], web search [16, 19], social choices [33, 36, 38, 39, 47], psychology [13, 34, 52], information retrieval [8, 32], etc.”
Spectral Method and Regularized MLE Are Both Optimal for Top-$K$ Ranking (2017, neutral): “This problem, which is called top-K rank aggregation, finds applications in numerous contexts, including web search (Dwork et al., 2001), recommendation systems (Baltrunas et al., 2010; Park et al., 2015), sports competition (Masse, 1997), and so on.”
Network medicine framework for identifying drug-repurposing opportunities for COVID-19 (2020, methodology): “As the optimal outcome, called the Kemeny consensus (29), is NP -hard to compute, we implemented three heuristic rank aggregation algorithms that approximate the Kemeny consensus: Borda ’ s count (30), the Dowdall method (31), and CRank (16).”
Comparing top k lists (2003, methodology): “, Tr (either full lists or top k lists) of "candidates," the rank aggregation problem [8] with respect to a distance measure d is to compute a list T (again, either a full list on the union of the domains of”
A feature selection model based on genetic rank aggregation for text sentiment classification (2017, background): “The aggregation of several different feature rankings may be beneficial to obtain an enhanced ranking of the features [38].”
Recommender Systems (2016, neutral): “This problem is that of rank aggregation [190].”
A large-scale evaluation and analysis of personalized search strategies (2007, methodology): “(3) Combine the rankings in τ1 and τ2 using Borda’ ranking fusion method [13, 8] and sort the web pages with the combined rankings.”
Handbook of Computational Social Choice (2016, background): “, Ephrati and Rosenschein, 1993), in the design of metasearch engines and in spam detection (e.g., Cohen et al., 1999; Dwork et al., 2001), in collaborative filtering and recommender systems (e.”
Measuring and Predicting Object Importance (2011, methodology): “This distance has already been applied in machine learning (Dwork et al. 2001; Lebanon and Lafferty 2002) to compare ranked lists.”
Bayesian inference for Plackett-Luce ranking models (2009, background): “Sparse data within the context of ranking is a common scenario for some applications and is typi.ed by having a small number of observations and a large number of items to rank (Dwork et al., 2001), or each individual observation may rank only a few of the total items.”