Even though plenty of researches have been issued in the field of association rules mining, there is still not a universal standard to judge the practicality of rules. Meanwhile, retailers often complain that they do not know how to optimize their businesses by the use of association rules. It seems as if the association rules can not play positive roles in decision-making processes. The topic of this dissertation is to realize the behavior of customers by analyzing retail transaction databases. In a certain extent, it provides some guidance in decision-making processes for retail enterprisers. It evolves certain significance for further researches in this field.The main contributions of this dissertation are as follows:(1) Using the traditional association rules mining algorithms have not yielded satisfactory results in the field of retailing. In the face of the weakness, this dissertation presents Matrix-FP-growth algorithm, an improvement of the FP-growth algorithm. It introduces a matrix to generate... |