Analysis of The Selection of The Best Arabica Coffee Beans Using Apriori Algorithms

Abstract

One of the coffee producers in North Sumatra is Karo Regency. Currently, the growth of coffee production in Indonesia is still hampered by the low quality of the coffee beans produced, which affects the development of the final coffee production. coffee, both in terms of selecting the best coffee beans and in terms of processing and benefits. To overcome this problem requires a system that can assist in making decisions regarding the selection of the best Arabica coffee beans with association pattern rules. As for applying associative rules to the selection of the best Arabica coffee beans in Tanah Karo between item combinations so as to form an itemset combination pattern and the Apriori Algorithm. Important association rules can be known with 2 parameters, namely, minimum support (the percentage of item combinations in the database) and minimum confidence (the strength of the relationship between items in the associative rules), both of which are determined by the user. The results of data mining calculations using the Apriori algorithm, data on the assessment of Mr. RM's coffee beans with a minimum support of 25% and a minimum confidence of 75%, form seven rules for Mr. RM's coffee beans. One of the best rules is that if the selection of Normal Bean Presentation is on the Andungsari 1, Andungsari 2, and Komasti coffee varieties, then with a 100% probability the selection on Leaf Rust Resistance will also be good for the selection of Andungsari 1, Andungsari 2, and Komasti coffee varieties. The best selection combination is {BN,B,KD,A} with a 100% confidence level and has the highest lift value, namely, 3.70.