Data Mining Analysis of Commodity Distribution in Central Aceh Through an Integrated Auction Market System Using the Android-Based Association Rule Mining Method

Commodity distribution in Central Aceh faces inefficiencies due to lengthy distribution chains and limited price control, which often leads to higher costs for consumers and lower profits for farmers. To address these issues, this study develops an integrated auction market system based on Android, utilizing the Association Rule Mining (ARM) method to optimize the distribution and pricing of commodities. ARM is a data mining technique that uncovers high-frequency patterns in transaction data. By applying ARM with the apriori algorithm, the system identifies key associations among commodities, allowing for more efficient and targeted price recommendations. The system calculates the highest bid for each commodity and recommends optimal pricing strategies to sellers based on frequent pattern analysis, improving transparency and reducing distribution inefficiencies. Testing and implementation of this system indicate that it successfully reduces distribution costs while increasing the effectiveness and speed of the auction process. Overall, the Android-based auction market system shows promise as a tool for enhancing distribution efficiency, optimizing bid values, and supporting local economies in Central Aceh through more equitable commodity pricing. The final result of this process resulted in four association rules based on predefined parameters, namely a minimum support of 20% and a minimum confidence of 50%. These rules indicate that 60% of the transactions in the Integrated Auction Market system that include Cassava also include Carrots. In other words, a bidder who buys Cassava has a 60% probability of also buying Carrots. This rule is significant as it shows that 20% of all transactions recorded in the system contain both items. This analysis provides important insights into the relationship patterns between items that can be used to provide item recommendations based on purchasing patterns.