Abstract
This study implements the Apriori algorithm-based Market Basket Analysis (MBA) on the Online Retail II dataset sourced from Kaggle, which contains 1,067,371 transaction records from a UK-based online retail company covering the period 2009–2011. The research aims to identify customer purchasing patterns and discover product combinations frequently bought together to support bundling strategies and cross-selling recommendations. After a data cleaning process that reduced the dataset to 776,872 valid records, a basket matrix was formed comprising 29,527 transactions and 120 top products. The Apriori algorithm was applied with a minimum support threshold of 2%, minimum confidence of 30%, and minimum lift of 1.0, resulting in 148 frequent itemsets and 44 association rules. The analysis revealed several strong product pair associations, including Roses Regency Teacup and Saucer and Green Regency Teacup and Saucer (support: 0.0252, confidence: 0.7036, lift: 22.2670), Spaceboy Lunch Box and Dolly Girl Lunch Box (confidence: 0.6065, lift: 17.9440), and Alarm Clock Bakelike Red and Alarm Clock Bakelike Green (confidence: 0.6136, lift: 17.4052). These findings provide valuable insights for retailers to design effective product bundling, optimize store layouts, and implement targeted promotional campaigns.
Open public full textRights: ALLOWED · cc-by-sa · Source: journal-auto-sync:CROSSREF_ISSN