Implementation of the Apriori Algorithm on Outdoor Equipment Rental Transaction Data Based on Clustering Using the K-Means Algorithm

Outdoor equipment rental services play a critical role in meeting climbers’ needs prior to expeditions. Sustaining business continuity in this sector requires effective marketing strategies, particularly given the increasing market competition. This study employs data mining techniques to analyze rental transaction data and identify patterns that support strategic decision-making. Specifically, clustering is performed using the K-Means algorithm to group transactions with similar attributes, followed by association rule mining using the Apriori algorithm within each cluster. A dataset comprising 1,276 valid transactions was processed, resulting in three clusters containing 324, 264, and 688 records, respectively, with an accuracy of 0.998. Apriori analysis generated 13 association rules in Cluster 0 and 2 rules in Cluster 1, while no rules met the minimum support and confidence thresholds in Cluster 2 or the overall dataset. These findings demonstrate that clustering prior to association rule mining can uncover meaningful patterns that are not evident in aggregated data. Such insights can inform targeted marketing strategies, including recommendations for item combinations frequently rented together. Future research may integrate alternative algorithms such as ECLAT or FP-Growth and explore framework-based systems to enhance scalability and precision in data-driven decision-making.