Joaquín Sánchez-Soriano

High dimensionality remains a significant challenge in dataset analysis, often leading to degraded classifier performance. To address this, feature selection has emerged as one of the most effective strategies. This paper presents a comparative computational study between game theory-based feature selection—specifically using Shapley and Banzhaf values—and several established algorithms implemented in the Weka workbench. The proposed methodology involves defining a cooperative game associated with a classification problem to calculate the Shapley and Banzhaf values for each feature; those with the highest values are selected based on their contribution to predictive accuracy. The performance of these game-theoretic methods was benchmarked against seven standard ranking algorithms across various datasets. Experimental results demonstrate that both Shapley and Banzhaf values effectively identify the most relevant features, performing competitively with traditional rankers. Consequently, this study suggests that game-theoretic approaches are robust and viable alternatives for feature selection in classification tasks.

In this paper, we extend the museum pass problem to incorporate the market structure. To be more precise, we consider that museums are organized into several pass programs or consortia. Within this framework, we propose four allocation mechanisms based on the market structure and the principles of proportionality and egalitarianism. Each mechanism satisfies a distinct set of reasonable properties related to fairness and stability, which serve to axiomatically characterize them.

The consumption of online videos is a rapidly expanding market that generates substantial revenue annually. A key challenge within this ecosystem is the equitable distribution of income between platform operators and content creators. This paper addresses this crucial problem by introducing a dynamic model of revenue generation that incorporates two primary income streams: pay-per-view systems and advertising insertion. To determine a fair and reasonable division of earnings, we model the interaction as dynamic cooperative games that reflect each party's contribution to value creation. We derive the structure of the Shapley value for these games and propose several related allocation rules. Furthermore, we provide an algorithmic framework for calculating these values and prove that the computational complexity is polynomial. We conclude with illustrative examples and simulations demonstrating the performance of the proposed allocation rules.

This paper presents a novel approach to variable selection in small area estimation, focusing on the Fay–Herriot model. Traditional methods, such as those based on Akaike and Kullback symmetric divergence information criteria, often rely on stepwise selection and focus on the complete model, without individually examining the influence of auxiliary variables. The Shapley value of cooperative game theory is proposed to measure the average importance of auxiliary variables. The Shapley value evaluates all possible combinations of auxiliary variables, ensuring an efficient average influence of the selected combination. We study its properties mathematically and investigate its performance through simulation experiments, showing consistent identification of the true generating variables even under challenging conditions. An application to the 2022 Spanish Living Conditions Survey illustrates the method’s usefulness in selecting the model on which to base predictors of poverty proportions in Spanish provinces by sex.

This paper proposes a novel algorithm to approximate the core of transferable utility (TU) cooperative games via linear programming. Given the computational hardness of determining the full core, our approach provides a tractable approximation by sampling extreme points through randomized linear problems (LPs). We analyze its convergence and computational complexity, and validate its effectiveness through extensive simulations on various game models. Our results show that the method is scalable and achieves high accuracy in terms of core reconstruction.

In this paper, we analyze the problem of how to adapt the concept of priority to situations where several perfectly divisible resources have to be allocated among a certain set of agents that have exactly one claim that is used for all resources. In particular, we introduce constrained sequential priority rules and two constrained random arrival rules which extend the classical sequential priority rules and the random arrival rule to these situations. We also provide an axiomatic analysis of these rules. Finally, we present a numerical example to compare the constrained random arrival rule to other solutions in this context.