Análise de desempenho de soluções para o problema de Máxima Interseção de k-Subconjuntos

Abstract

The Maximum Intersection of k Subsets problem involves finding the maximum number of elements present in the intersection of k subsets, where k is a positive integer. In this sense, this problem is applied as a way to abstract real-world problems, such as hiding sensitive patient information when publishing scientific research, in genetic research to find patterns in genetic data sets, or in the context of social networks to classify groups of users with similar characteristics. In this work, the objective is to perform a comparative analysis between the main solutions to this problem, exploring the best heuristic strategies, specifically using Ant Colony Optimization (ACO), GRASP (Greedy Randomized Adaptive Search Procedure), and VNS (Variable Neighborhood Search), instantiating tests from a real data set of Amazon movie reviews, seeking to find the best solutions for the context of social networks.