Implementation of Data Mining Using Apriori and FP-Growth Algorithms to Determine Product Bundling Strategy in Sakinah Supermarket Keputih

The FMCG industry in Indonesia experienced a slowdown in growth in 2023 due to national inflation, which altered consumer spending patterns. Sakinah Supermarket, a leading player in the sector, also saw a revenue decline attributed to global inflation. To address this, Sakinah plans to develop a new promotional strategy based on consumer shopping behaviours. One proposed approach is using market basket analysis to create product bundling strategies. This involves analysing transaction data through Data Mining using the Apriori and FP-Growth algorithms, which efficiently identify frequent item sets and generate association rules. The dataset used in this study includes 3634 transactions and 22,413 rows of data, which are the data transactions from January 2023 until december 2023. The algorithms will be compared based on support, confidence, and lift ratio values to determine the most accurate method. The selected algorithm's results will guide product bundling recommendations for Sakinah Supermarket. Based on the percentage value of the data processing results using the apriori and FP-Growth algorithms, it is possible to decide that Sakinah Supermarket's data is better suited to the FP-Growth approach. The resulting item combinations are Sedaap Mie and Indomie Inst with a lift ratio value of 3,7; Sunlight SBN Cuci Prg and Indomie Inst with a lift ratio value of 2,5; Soklin Softgn and Indomie Inst with a lift ratio value of 2,2; Ultra Milk Uht and Indomie Inst with a lift ratio value of 1,7; and Pepsodent PG and Indomie Inst with a lift ratio value ofl,6. All combinations show a lift value greater than 1, suggesting that the resulting combination is legitimate and suitable for bundling product items at Sakinah Supermarket.