Implementasi Data Mining Penerimaan Mahasiswa Baru Universitas Graha Karya Muara Bulian

Abstract

The choice of prospective students on the registration form produces a unique pattern because each prospective student has a different choice according to background, aspirations and tuition fees while still referring to the 9 courses offered. Piles of forms with combinations of program study options will have no meaning for UGK MB if detailed mining and analysis is not carried out. In fact, doing useful data mining will gain knowledge and information that is useful for UGK MB and other decision makers to address the process of admitting new students in the following year. Difficulties in answering past data stacks, which are often also known as chunks of data of this kind, can be ported with data mining (Data Mining). The amount of data studied is 135 forms. The method used is Apriori Algorithm analysis. The results show that for 1-itemset, the highest frequency occurs in the Information Systems variable with a value of 58, 43% support and 100% confidence, while the lowest frequency occurs in Accounting with a value of 14, 10% support and 100% confidence. For the 2-item set, the association rules If the first choice is the PGSD study program then the second choice is the TIP study program with a frequency value of 16, 12% support and 53% confidence and this is the highest association rule. Meanwhile, the lowest association rule is the rule. If the first choice is the Agro Technology study program, then the second choice is the Information Systems study program with a frequency of 4, 3% support and 20% confidence.