<![CDATA[Abstrak - Industri kuliner di Indonesia, khususnya jaringan kedai es krim Mixue, menghadapi tantangan dalam memberikan rekomendasi menu yang tepat kepada pelanggan. Dengan beragamnya pilihan menu dan preferensi individu yang berbeda, sistem rekomendasi yang efektif sangat diperlukan untuk meningkatkan pengalaman pelanggan dan penjualan. Penelitian ini bertujuan untuk mengembangkan sistem rekomendasi makanan di Mixue menggunakan metode collaborative filtering, khususnya pendekatan item-based collaborative filtering, yang memanfaatkan data transaksi pelanggan untuk memberikan saran menu yang relevan. Metodologi yang digunakan mencakup pengumpulan data transaksi pelanggan, perhitungan kesamaan antar item menggunakan metrik cosine similarity, dan pembangunan model rekomendasi berbasis algoritma k-nearest neighbors (k-NN). Sistem ini diintegrasikan dengan aplikasi Mixue,memungkinkan rekomendasi menu ditampilkan secara real-time kepada pengguna.Kata Kunci: Mixue; Sistem rekomendasi; Collaborative filtering; Item-based; k-Nearest Neighbors. Abstract - The culinary industry in Indonesia, particularly the Mixue ice cream shop network, faces challenges in providing the right menu recommendations to customers. With a variety of menu options and different individual preferences, an effective recommendation system is essential to enhance customer experience and sales. This study aims to develop a food recommendation system at Mixue using the collaborative filtering method, specifically an item-based collaborative filtering approach, which utilizes customer transaction data to provide relevant menu suggestions. The methodology used includes collecting customer transaction data, calculating item similarities using the cosine similarity metric, and building a recommendation model based on the k-nearest neighbors (k-NN) algorithm. This system is integrated with the Mixue application database, allowing menu recommendations to be displayed in real-time to users.Keywords: Mixue; Recommendation system; Collaborative filtering; Item-based; k-Nearest Neighbors.]]>
