Basket Purchase Prediction Using Association Rule Mining Based on the FP-Growth Algorithm

Market basket analysis helps marketing analysts understand customer behavior such as identifying which products are frequently purchased together. Various data mining techniques and algorithms have been developed to perform such analyses. The present study introduces an innovative approach that applies the FP-Growth algorithm to discover associations among users’ purchases in order to enhance the efficiency of e-commerce systems. In the proposed method, all user transactions are utilized in the basket analysis process. In other words, even purchases that appear unrelated to the user’s current transactions can provide valuable information, contributing to a deeper understanding of customer purchase patterns and improving the overall performance of sales systems. To evaluate the effectiveness of the proposed method, its results were compared with those of the Eclat and Apriori algorithms. Experimental analyses revealed that, on average, the proposed method outperformed the compared approaches.

کلیدواژه ها:

Data mining ، Basket Purchase Prediction ، Association Rule Discovery ، Frequent Pattern Growth