<![CDATA[The rapid expansion of the takeaway food industry through digital platforms has generated a vast volume of transaction data. A significant challenge lies in efficiently extracting valuable insights from large databases without overtaxing computational memory. This research proposed the implementation of the Frequent Pattern Growth (FP-Growth) algorithm to map association patterns between food items without the need for iterative candidate itemset generation. The experiment was conducted on the Takeaway Food Orders dataset from an Indian restaurant in London, comprising 13,397 transactions and 74,818 data rows. The methodology involved data pre-processing, FP-Tree construction, and the determination of strong association rules. Results demonstrated that the algorithm effectively identified customer ordering trends. By applying a 3% support threshold and 75% confidence, seven significant association rules were identified, with Plain Papadum serving as the consequent in all rules. These rules indicate that Plain Papadum has a high tendency to be ordered together with Mint Sauce, Onion Chutney, Mango Chutney, Pilau Rice, and Red Sauce. This study concluded that FP-Growth-based data mining provides accurate strategic insights for business owners to optimize promotional bundling and enhance customer loyalty.]]>
Copyrights © 2026
