Incremental Granular Ranking Algorithm Based on Rough Sets

Based on the original granular ranking algorithm proposed by the author, this paper proposes a granularity set combination algorithm with the time complexity of O(mn), thus constructs an incremental granular ranking algorithm. The computation result of new algorithm is in the form of ranked list, which can avoid many disadvantages in traditional data mining techniques, and can be applied for the investigation of targeted marketing in large-scale datasets, identifying potential market values of customers or products. The experiment result also shows that the accuracy of computation result can be improved obviously by adding new training datasets.