Beyond Pairwise Comparisons in Social Choice: A Setwise Kemeny Aggregation Problem

Authors: Hugo Gilbert, Tom Portoleau, Olivier Spanjaard1982-1989

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our numerical tests have three objectives: we evaluate the computational performance of the dynamic programming approach of Section 3, we evaluate the impact of parameter k on the set of consensus rankings, and we assess the efficiency of the preprocessing technique of Section 4.
Researcher Affiliation Academia Hugo Gilbert Gran Sasso Science Institute 67100 L Aquila, Italy hugo.gilbert@gssi.it Tom Portoleau LAAS-CNRS, IRIT-CNRS Universit e de Toulouse 31400 Toulouse, France tom.portoleau@laas.fr Olivier Spanjaard Sorbonne Universit e CNRS, LIP6, 75005 Paris, France olivier.spanjaard@lip6.fr
Pseudocode No The paper describes algorithms and mathematical formulations but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access information for open-source code for the methodology described.
Open Datasets No The paper states that preference profiles are 'generated according to the Mallows model' using a 'Python package Pref Lib-Tools'. This indicates data generation rather than the use of a pre-existing publicly available dataset with a specific link or citation.
Dataset Splits No The paper describes data generation and sets 'the number n of voters to 50' and varies 'm, k, φ'. However, it does not specify any explicit training, validation, or test dataset splits, or refer to standard benchmark splits.
Hardware Specification Yes All times are CPU seconds on an Intel Core I7-8700 3.20 GHz processor with 16GB of RAM.
Software Dependencies No The paper mentions 'Implementation in C++' and 'using the Python package Pref Lib-Tools (Mattei and Walsh 2013)'. While Python and a package are named, specific version numbers for the package or any other libraries are not provided, which is necessary for a reproducible description.
Experiment Setup Yes The preference profiles are generated according to the Mallows model (Mallows 1957), using the Python package Pref Lib-Tools (Mattei and Walsh 2013). This model takes two parameters as input: a reference ranking σ (the mode of the distribution) and a dispersion parameter φ (0, 1). ... In all tests, the number n of voters is set to 50 and the ranking σ is set arbitrarily as the k-wise Kemeny rule is neutral. For each triple (m, k, φ) considered, the results are averaged over 50 preference profiles.