<![CDATA[In the digital era, transaction data analysis has become an important strategy for increasing business effectiveness, including bakeries. One technique that is often used is association analysis, with the Apriori and FP-Growth algorithms as the main approaches. The main problem is to determine a more efficient and relevant algorithm to support business decision making in bakeries, especially in purchasing pattern analysis. This research aims to compare the performance of the Apriori and FP-Growth algorithms in determining association rules in bakery transaction data. The research methodology includes transaction data collection, data pre-processing, algorithm implementation, performance testing, and results analysis. Parameters analyzed include processing time, number of generated rules, and efficiency at different dataset scales. Testing was carried out with Python software, utilizing libraries such as mlxtend and pandas. The research results show that the FP-Growth algorithm is superior in time efficiency and scalability, especially on large datasets, thanks to its approach that does not require the formation of candidate itemsets. Meanwhile, the Apriori algorithm is easier to implement and remains relevant for small datasets with the right parameters. This research offers novelty by focusing on the application of both algorithms in a bakery context, providing specific insights into how to optimize stock management and design data-driven promotions. These results provide an important contribution for bakeries and similar business sectors to utilize data analysis as a strategic tool in increasing customer satisfaction and competitiveness.]]>
