Understanding Distance Measures Among Elections

Authors: Niclas Boehmer, Piotr Faliszewski, Rolf Niedermeier, Stanisław Szufa, Tomasz Wąs

IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental While in the previous section we studied distances between hand-crafted elections, now we analyze automatically-generated ones. We test how our metrics correlate with the swap one, and we compare their maps of elections. We use two datasets. For each dataset we have computed the Pearson Correlation Coefficient (PCC) between the swap distances and those provided by the other metrics. PCC is a classic measure of correlation that takes values between 1 and 1; its absolute value gives the strength of the correlation and the sign indicates its positive or negative nature. Szufa et al. [2020] presented a similar experiment, but on a much smaller scale, and on a limited set of metrics. We present our results in Table 2.
Researcher Affiliation Academia 1Technische Universität Berlin, Algorithmics and Computational Complexity 2AGH University 3University of Warsaw
Pseudocode No The paper describes methods and proofs in prose, but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide an explicit statement about releasing its source code, nor does it include a link to a code repository for the described methodology.
Open Datasets Yes Briefly put, such a map is a collection of election instances, typically generated from some statistical models (but Boehmer et al. [2021b] also used real-life ones from Pref Lib [Mattei and Walsh, 2013]). We use two datasets. The first one consists of all small elections, as in Section 4. The second one resembles those used in the maps of Szufa et al. [2020] and Boehmer et al. [2021b], but consists of elections with 10 candidates and 50 voters,3 generated according to the following statistical models (see the just-cited papers for more details): IC, Urn, and Mallows.
Dataset Splits No The paper describes how datasets were generated and used for analysis, but does not discuss training, validation, or test splits, as its experimental approach does not involve training a model.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., programming languages, libraries, or solvers).
Experiment Setup No The paper describes how election instances were generated and refers to external algorithms for visualization, but it does not provide specific experimental setup details such as hyperparameter values, training configurations, or system-level settings.