OPTIMASI PENGGUNAAN ASOSIASI RULE DENGAN ALGORITMA APRIORI UNTUK ANALISA POLA TRANSAKSI PELANGGAN TELKOMSEL

The rapid development of information and communication technology has prompted telecommunications companies to improve service quality and gain a deeper understanding of customer behavior. Telkomsel is one such company with a large subscriber base in Indonesia. As the number of subscribers continues to grow, the volume and complexity of transaction data—particularly in the Kisaran region—have also increased. This stored transaction data holds significant potential for analysis to uncover customer behavioral patterns, such as tendencies toward purchasing data packages, mobile credit, or other digital services. The objective of this study is to apply the Apriori algorithm for association rule analysis on customer transaction data. The study utilizes historical sales transaction data from Telkomsel, including purchases of data packages, mobile credit, digital services, and SIM cards. The findings reveal association patterns (Rules 2 & 3): a combination of "Internet OMG" packages with "Kuota Ketengan" (short-term/specific-use data quotas) (Support 35%, Confidence 65%), and voice call packages paired with SMS packages (Support 30%, Confidence 60%), which offer potential options for add-on services. These patterns are highly suitable for use as a dynamic recommendation system on the transaction interface to boost cross-selling.