DATA MINING DENGAN ALGORITMA DYNAMICSOME UNTUK PENENTUAN PENGIRIMAN DAN STOK YANG BELUM DI KIRIM PUPUK SUBSIDI

Abstract

Data Mining is a new technology that is very useful to help companies find very important information from their data warehouses. Some data mining applications focus on prediction, they predict what will happen in a new situation from data that describes what happened in the past. In order to find out how many tons of fertilizer have been sent and how many have not been sent by the subsidized fertilizer factory appointed by the government. can be done by using analytical techniques, namely the analysis of consumer buying habits. Detection of fertilizers that are often purchased together is done using association rules. In this study, an a priori algorithm will be used to determine the association rules for buying fertilizer.

So in solving these problems, the DynamicSome Algorithm method is used, which is a modification of the Apriori algorithm which will search for frequent itemsets from transaction data. Frequent itemset is a pair of items found in transaction data. In addition, the DynamicSome algorithm is also used to analyze the relationship between different items in a large set of items, this aims to see the relationship and attachment between these items, from the calculation results obtained in the 2016 tax year there are still 5 districts that have not been sent and in 2017 still 3 districts have not been sent.