Drawing a Map of Elections
🤖 AI Summary
Existing frameworks lack geometric representations that systematically capture similarities among voting rules, preference structures, and social choice outcomes in a unified, interpretable space.
Method: We propose the “Election Map” paradigm—a two-dimensional Euclidean embedding where distances between points reflect behavioral similarity across voting rules and preference profiles. Our approach integrates computational social choice theory, multidimensional scaling (MDS), and kernel embedding techniques, validated on both real-world and synthetic election datasets.
Contribution/Results: This work provides the first visualization and quantitative distance metric for comparing voting rules, uncovering previously uncharacterized structural relationships beyond classical axiomatic analysis. Empirical evaluation demonstrates high consistency and robustness in predicting rule-level behavioral similarity. The Election Map thus offers a novel, interpretable tool for election mechanism design, rule evaluation, and institutional analysis—bridging theoretical social choice with data-driven empirical modeling.
Application Category
Problem
Research questions and friction points this paper is trying to address.
Introducing a framework to map elections with synthetic and real-life datasets
Measuring election similarities using polynomial-time computable positionwise distance
Visualizing elections in 2D space for analysis via coloring criteria
Innovation
Methods, ideas, or system contributions that make the work stand out.
Framework for mapping elections with synthetic datasets
Polynomial-time positionwise distance for similarity measurement
Kamada-Kawai algorithm for 2D election representation
🔎 Similar Papers
No similar papers found.
