Implementasi Data Mining terhadap Pola Pembelian Skincare menggunakan Metode Algoritma Apriori

Lolyta Erma (Universitas Bina Sarana Informatika)
Rahmia Puji (Universitas Bina Sarana Informatika)
Angelina Dexter (Universitas Bina Sarana Informatika)
Lia Ruhlia (Universitas Bina Sarana Informatika)
Nur Aini Setyawati (Universitas Bina Sarana Informatika)

Jurnal Nasional Komputasi dan Teknologi Informasi
Vol. 9 No. 4 (2026): Agustus, 2026

Abstrak - Penelitian ini bertujuan menerapkan algoritma Apriori untuk menemukan pola pembelian produk skincare. Masalah utamanya adalah data transaksi yang ada belum dimanfaatkan secara optimal untuk menyusun rekomendasi paket produk, mengatur stok, dan merancang strategi promosi. Data yang digunakan berupa 1024 transaksi sampel yang mencakup pembelian facial wash, toner, serum, moisturizer, sunscreen, micellar water, acne spot treatment, sheet mask, exfoliating toner, lip balm, eye cream, dan clay mask. Tahapan penelitian meliputi pembersihan data, pengkodean item, transformasi transaksi ke matriks biner, pembentukan frequent itemset, perhitungan support, confidence, dan lift ratio, serta interpretasi aturan asosiasi. Hasilnya menunjukkan bahwa kombinasi facial wash, toner, moisturizer, dan sunscreen jadi pola pembelian paling dominan. Aturan serum ke toner menghasilkan confidence 100% dengan support 41%, sementara aturan facial wash dan serum ke sunscreen menghasilkan confidence 100%, support 32%, dan lift ratio 1,54. Temuan ini bisa dijadikan dasar untuk rekomendasi bundling, pengaturan stok, dan promosi produk pelengkap. Kata kunci: Data Mining; Apriori; Skincare. Abstract - This study applied the Apriori algorithm to identify skincare purchasing patterns. The main problem was the limited use of transaction data for product bundle recommendations, inventory control, and promotional strategy. A total of 1024 sample transactions were used, representing purchases of facial wash, toner, serum, moisturizer, sunscreen, micellar water, acne spot treatment, sheet mask, exfoliating toner, lip balm, eye cream, and clay mask. The research process covered data cleaning, item coding, binary transaction transformation, frequent itemset generation, support, confidence, and lift ratio calculation, and association rule interpretation. Results showed that facial wash, toner, moisturizer, and sunscreen formed the most dominant purchasing pattern. The rule from serum to toner achieved 100% confidence with 41% support, while the rule from facial wash and serum to sunscreen achieved 100% confidence, 32% support, and a lift ratio of 1.54. These findings can support bundle recommendations, stock arrangement, and complementary product promotions. Keywords: Data Mining; Apriori; Skincare.