This paper presents a game-theoretical framework for data classification and network discovery, focusing on pairwise influences in multivariate choices. The framework consists of two complementary games in which individuals, connected through a signed weighted graph, exhibit network similarity. A voting rule captures the influence of an individual’s neighbors, categorized as attractive (friend-like) or repulsive (enemy-like), and encodes individuals’ payoffs based on network similarity. We establish a duality between these two games, distinguishing between the endogeneity of choices (direct voting) and network information (inverse voting). While the latter has applications in network discovery, the direct voting game results in a data classification methodology that generalizes the $K$-nearest neighbors approach. Our theoretical results provide conditions for the existence of Nash equilibria and demonstrate the NP-completeness of their characterization. On the empirical side, we test our methodology with three applications and show the advantages (in terms of goodness-of-fit) of our game-theoretical framework in addressing both data classification and network discovery.