A Hybrid GRASP with Data Mining for the Maximum Diversity Problem

The maximum diversity problem (MDP) consists in identifying, in a population, a subset of elements, characterized by a set of attributes, that present the most diverse characteristics among themselves. The identification of such solution is an NP-hard problem. In this work, we propose a hybrid GRASP metaheuristic for the MDP that incorporates a data mining process. Data mining refers to the extraction of new and potentially useful knowledge from datasets in terms of patterns and rules. We believe that data mining techniques can be used to extract patterns that represent characteristics of sub-optimal solutions of a combinatorial optimization problem. Therefore these patterns can be used to guide the search for better solutions in metaheuristics procedures. Performance comparison between related work and the proposed hybrid heuristics is provided. Experimental results show that the new hybrid GRASP is quite robust and, mainly, this strategy is able to find high-quality solutions in less computational time.