Big Data Mining Method of New Retail Economy Based on Association Rules

Abstract

There are many sparse items in the data of new retail industry, and the extracted association rules are redundant, which leads to the problems of low temporal and spatial efficiency and poor mining quality when applied to the data mining of new retail economy. In order to improve the quality of data mining, a new retail economy big data mining method based on association rules is proposed. The k-means algorithm is used to subdivide the customer groups under the new retail economic model and extract the corresponding data association rules. Introduce interest threshold to filter association rules. The weight of frequent set is introduced to establish frequent item tree, and FP-Growth algorithm is improved to realize big data mining of new retail economy. Simulation results show that when the proposed data mining method is applied to the big data processing of the new retail economy, the execution efficiency and execution quality are improved by at least 27.5%.

Implementation of Association Rules Using Apripori Algorithm for Angkringan

“…In the Apriori algorithm it is known as support and confidence. The formula for support and confidence is as below (Liu, 2023):…”

“…An example of an association rule from purchasing analysis in a supermarket is knowing how likely it is that a customer buys coffee together with milk. With this knowledge, supermarket owners can arrange the placement of their goods or design marketing campaigns using discount coupons for certain product combinations (Liu, 2023), (Siva et al, 2019). Association analysis became famous because of its application to analyze the contents of shopping baskets in supermarkets.…”