<![CDATA[The problem that occurs in increasing sales transactions is that the large number of sales transactions every day with many kinds of spare parts makes it difficult for sales to determine strategies for offering spare parts that are relevant and really needed by consumers. The large amount of transaction data to be analyzed is not possible to do manually. Therefore, a certain technique is needed that can carry out the association rule mining process quickly on quite large data. One technique that can be used for association rules is the Hash algorithm. Hash Based Algorithm Uses hashing techniques to filter out itemsets that are not important for generating the next itemset. Generating frequent itemsets in this research requires a minimum support value and a minimum confidence value. The minimum support value used in this research is the average frequency of spare parts sales. Development is carried out in the process of generating candidate rules using a hash table on transaction data. By using the Hash Based algorithm, the association rule mining process becomes faster and more efficient in memory usage. From the results of the association rules trial with a minimum support of 40% and a minimum confidence of 75%. 14 lists of association rules were produced that met the requirements. This list becomes a reference for sales in determining strategies for offering spare parts to consumers.]]>
Copyrights © 2024
