A0413
Title: A generalized voting game for categorical network choices
Authors: Yueh Lin - IESEG School of Management (France) [presenting]
Stefano Nasini - IESEG School of Management (France)
Martine Labbe - Free University of Bruxelles (Belgium)
Abstract: A game theoretical framework is presented for data classification based on the interplay of pairwise influences in multivariate choices. This consists of a voting game wherein individuals, connected through a weighted network, select features from a finite list. A voting rule captures the positive or negative influence of an individual's neighbours, categorized as attractive (friend-like relationships) or repulsive (enemy-like relationships). Payoffs are assigned based on the total number of matching choices from an individual's neighbours. It is shown that the approach constitutes a natural generalization of the K-nearest neighbors method, establishing the proposed game as a theoretical framework for data classification. Computationally, a mixed-integer linear programming formulation is constructed to approach the Nash equilibria of the game, facilitating their applicability to real-world data. The results provide conditions for the existence of Nash equilibria and for the NP-completeness of its characterization. On the empirical side, the proposed approach is used to impute missing data and highlight its competitive advantage over the K-nearest neighbors approach.
