A paper on a hybrid algorithm turned out to be a hybrid itself — some original data, plus some from a paper that the authors had published earlier.
According to the retraction note, the overlap was significant enough to pull it from the scientific record.
The retracted paper describes an algorithm that is the combination of a “genetic algorithm” and a “cultural algorithm”– which, as their names sort of suggest, focus on looking at a population of solutions, and the history of which kinds of solutions work, respectively. According to the abstract, results to optimization problems found with a hybrid algorithm are “more accurate and the fast convergence is obvious.”
The retraction note provides a few details about the nature of the duplication: