Implementasi Data Mining dengan Metode Apriori Dalam Menentukan Pola Pemilihan Pemeriksaan Kimia

Abstract

Businesses and organizations that want to survive need to develop effective business strategies. Reviewing the company's transaction data gradually leads to sales data. So it is very if not re-analyzed. The examination products offered at Prodia Laboratories vary, and the exploration of the packages offered influences the public to choose these packages. The Apriori method is a method that can help companies find out which test inspections are the most sold and know the relationship between inspections with one another, and with the help of the RapidMiner software the types of tests or tests that come out at the same time can be known. The a priori algorithm is a type of association rule in data mining (data mining). One of the stages of Association Analysis that has attracted the attention of many researchers to produce efficient algorithms is high-frequency pattern mining. The importance or not of an association rule can be understood with two benchmarks, namely the support value and the confidence value. Support (support value) is the proportion of the combination of these items in the database, while confidence (value) is the strength of the relationship between items in the association rules. From the results of calculations using the Apriori method to determine purchasing patterns on the type of chemical/fat check by determining a minimum support of 35% and 75% confidence, resulting in 3 association rules, namely first, if you buy Total Cholesterol, you will buy LDL Cholesterol with 100% support and 100% confidence, secondly. If Total Cholesterol, then I will buy Triglycerides with 100% support and 100% confidence, buy all three. If I buy LDL Cholesterol, I will buy Triglycerides with 100% support and 100% confidence