Skripsi/Tugas Akhir
Perancangan Sistem Pengelompokan dan Rekomendasi Buku Rumah Baca Saku Menggunakan Algoritma Hierarchical Clustering Berbasis Web
ABSTRAK
Rumah Baca Saku menghadapi kendala dalam pengelolaan koleksi buku, di mana buku-buku belum dikelompokkan berdasarkan kategori, sehingga pengunjung kesulitan menemukan buku sesuai minat mereka. Selain itu, proses rekomendasi masih dilakukan secara manual, yang memerlukan waktu cukup lama. Untuk mengatasi permasalahan ini, dikembangkan sistem rekomendasi buku berbasis web menggunakan metode Hierarchical Clustering dan Content-Based Filtering, Pada penelitian ini dilakukan tahapan yang mencakup pengumpulan data sebanyak 220 buku , kemudian Transformasi data menggunakan One-Hot Encoding perhitungan jarak menggunakan Euclidean Distance, serta metode Single Linkage untuk menggabungkan buku dengan jarak terdekat hingga membentuk dendogram. Berdasarkan hasil pemotongan dendogram pada jarak 20, diperoleh lima klaster, yaitu klaster Fiksi, Non Fiksi, Referensi, Pendidikan, dan Buku Anak. Pada rekomendasi buku Pengunjung dapat melihat daftar buku berdasarkan kategori yang diminati, serta Informasi seperti judul, penulis, penerbit, tahun terbit, ISBN, dan deskripsi buku. Pengujian Black Box Testing menunjukkan sistem berfungsi dengan sesuai yang diharapkan, memudahkan pengunjung dalam pencarian buku dan membantu admin mengelola koleksi secara lebih terstruktur.
Kata kunci: Rekomendasi, Pengelompokkan, Buku, dan Hierarchical Clustering
ABSTRACT
Rumah Baca Saku faces obstacles in managing book collections, where books have not been grouped by category, so visitors have difficulty finding books according to their interests. In addition, the recommendation process is still done manually, which takes quite a long time. To overcome this problem, a web-based book recommendation system was developed using the Hierarchical Clustering and Content-Based Filtering methods. In this study, stages were carried out that included collecting data on 220 books, then transforming data using One-Hot Encoding, calculating distances using Euclidean Distance, and the Single Linkage method to combine books with the closest distance to form a dendogram. Based on the results of cutting the dendogram at a distance of 20, five clusters were obtained, namely the Fiction, Non-Fiction, Reference, Education, and Children's Book clusters. In the book recommendations, visitors can see a list of books based on the category they are interested in, as well as information such as title, author, publisher, year of publication, ISBN, and book description. Black Box Testing shows that the system functions as expected, making it easier for visitors to search for books and helping admins manage collections in a more structured manner.
Keywords: Recommendations, Grouping, Books, and Hierarchical Clustering
Tidak ada salinan data
Universitas DIPA Makassar
NPP 7371142D1000002
Jln. Perintis Kemerdekaan KM.9
Telp. (0411)587194
Hotline: +6281228221994
WhatsApp Admin: +6281342092072
e-Mail: perpustakaan@undipa.ac.id
© 2025 — Perpustakaan UNDIPA Makassar - SLiMS