Skip to Main Content
With the market competition being increasingly intensive, it is necessary for company to carry out one-to-one marketing to consumers. Therefore, the ability to predict consumer's consumption behavior basing on data mining has become a key source of competitive advantage for company. In this paper, we propose a novel algorithm, which bases on ant colony optimization (ACO) to cluster consumer's consumption data sets with different favor for predicting consumer's behavior. The algorithm is applied to grouping consumer data into classes or clusters for revealing consumer's consumption behavior. The novel algorithm adopts simulated annealing concept for ants to decreasingly visit cities to get local optimal solutions. Finally, the algorithm is validated by the example of clustering consumer's credit card data. The result indicates the algorithm is successful in clustering data for analyzing consumer's behavior. The research presented in this paper makes contribution to predicting consumer's consumption behavior.