An orderly algorithm for generation of Condorcet Domains
By: Bei Zhou, Klas Markström
Potential Business Impact:
Finds best voting rules for fair elections.
Condorcet domains are fundamental objects in the theory of majority voting; they are sets of linear orders with the property that if every voter picks a linear order from this set, assuming that the number of voters is odd, and alternatives are ranked according to the pairwise majority ranking, then the result is a linear order on the set of all alternatives. In this paper we present an efficient orderly algorithm for the generation of all non-isomorphic maximal Condorcet domains on $n$ alternatives. The algorithm can be adapted to generate domains from various important subclasses of Condorcet domains. We use an example implementation to extend existing enumerations of domains from several such subclasses and make both data and the implementation publicly available.
Similar Papers
Improved lower bounds for the maximum size of Condorcet domains
Discrete Mathematics
Finds better ways to count votes fairly.
Approximately Dominating Sets in Elections
CS and Game Theory
Finds a few good winners even in tricky elections.
Pairwise similarity method for majority domination problem
Discrete Mathematics
Finds how many votes guarantee a win.