Analysis of Agricultural Product Package Recommendations Using the FP-Growth Algorithm

Abstract

The size of the agricultural sector in Indonesia provides opportunities for small businesses managed by local residents to provide various needs to support agricultural activities. The rise of agricultural shops has made business competition in the agricultural sector quite competitive. Therefore, a marketing strategy is needed that can increase the sales of various marketed agricultural products. The method applied in this research is Association Rule Mining using the FP-Growth algorithm. This method is applied to obtain association rules that are used as a reference in making recommendations for agricultural product packages at agricultural shops to increase sales. The system workflow starts from preprocessing data to form an itemset which is then processed using the FP-Growth algorithm. The next step is to determine the minimum value of support and minimum confidence as a limit in calculating the FP-Growth algorithm. The system will eliminate a number of itemsets that do not meet the specified threshold to produce frequent itemsets which are then mined into rules as a reference to form the most recommended package of agricultural products. From the research conducted it can be seen that there are 3 itemets that almost always appear and the most recommended agricultural product package is Prowl 250 ml (Herbicide) which is associated with Antracol 70wp 1 kg (fungicide) with a support value of 6.38% and a confidence value of 85, 71% and the lift ratio is 8.95.