A Data Mining and Processing Method for E-Commerce Potential Customers Based on Apriori Association Rules Algorithm

In order to improve the effectiveness of e-commerce potential customer data mining and processing, a method based on Apriori association rule algorithm for e-commerce potential customer data mining and processing is proposed. Innovatively adopting a multidimensional tree structure to improve the Apriori association rule algorithm, using frequent itemsets as candidate itemsets, and further expanding on this basis by adding judgment conditions to reduce the frequency of scanning the database; The Vector space model is used to calculate the similarity between e-commerce potential customers, and the similarity is used as a scalar value to complete the accurate calculation. The e-commerce potential customers at different levels in customer transaction data are divided. Obtain a sticky evaluation system for potential e-commerce customers from the perspectives of perceived usefulness, perceived ease of use, perceived service, perceived security, and perceived interest, as the basic indicators for subsequent mining and processing. The Quicksort method is used to sort each data dimension in the e-commerce customer data set, and the improved Apriori association rule algorithm is used to realize data mining and processing of e-commerce potential customers through high-density grid. The experimental results demonstrate that the method innovatively utilizes the improved Apriori association rule algorithm to mine three types of customer behavior data with an accuracy of over 80%, which is in line with the actual situation. It improves the effectiveness of e-commerce potential customer data mining and processing, effectively mining e-commerce potential customers, and providing good basic data for e-commerce platforms to adjust marketing strategies.