| dc.contributor.advisor | Indriasari, Sofiyanti | |
| dc.contributor.author | KURNIAWAN, MAHESA DZIKRI | |
| dc.date.accessioned | 2026-08-12T04:29:39Z | |
| dc.date.available | 2026-08-12T04:29:39Z | |
| dc.date.issued | 2026 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/178390 | |
| dc.description.abstract | Ledakan volume konten komik digital memicu masalah penumpukan informasi (information overload) yang mempersulit pembaca menemukan konten yang sesuai. Selain itu, sistem rekomendasi konvensional sering kali menghadapi kendala Cold Start dan bias popularitas. Penelitian ini bertujuan merancang dan membangun website komik fungsional yang dilengkapi sistem rekomendasi berbasis aturan (Rule-Based System) dan statistika deskriptif untuk mengatasi tantangan tersebut. Metode pengembangan sistem menggunakan prosedur Agile Kanban. Sistem ini mengimplementasikan fitur Interest Picker untuk menangkap preferensi awal pengguna baru guna mengatasi masalah Cold Start. Untuk pembaca aktif, sistem merekam perilaku membaca implisit (durasi baca dan kedalaman scroll) secara otomatis, menyaring anomali interaksi melalui Behavioral Negative Filter, serta mengekstraksi lima genre dominan teratas untuk menghasilkan rekomendasi personal. Fitur gamifikasi gelar juga diterapkan untuk meningkatkan keterlibatan pengguna. Hasil pengujian fungsionalitas dengan Black Box Testing mencapai persentase keberhasilan sebesar 98,81%. Evaluasi User Acceptance Testing menunjukkan tingkat penerimaan sangat baik pada semua aspek, dengan skor kenyamanan mencapai 90,32% dan keterlibatan pengguna sebesar 89,92%. Penelitian ini membuktikan bahwa kombinasi umpan balik implisit dan logika berbasis aturan mampu menyajikan rekomendasi komik yang relevan dan meningkatkan keterikatan pembaca secara signifikan. | |
| dc.description.abstract | The explosion of digital comic content volume triggers information overload, making it difficult for readers to find suitable content. Moreover, conventional recommendation systems often face Cold Start and popularity bias challenges. This study aims to design and develop a functional comic website equipped with a recommendation system based on a Rule-Based System and descriptive statistics to address these issues. The system development uses the Agile Kanban methodology. The system implements an Interest Picker feature to capture new users' initial preferences to overcome the Cold Start problem. For active readers, the system automatically records implicit reading behavior (reading duration and scroll depth), filters interaction anomalies through a Behavioral Negative Filter, and extracts the top five dominant genres to generate personalized recommendations. A title achievement gamification feature is also applied to enhance user engagement. Functional testing using Black Box Testing achieved a success rate of 98.81%. User Acceptance Testing evaluations showed excellent acceptance across all variables, with satisfaction scoring 90.32% and user engagement at 89.92%. This study demonstrates that combining implicit feedback and rule-based logic effectively provides relevant comic recommendations and significantly increases reader engagement. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Pembuatan Website Komik dengan Sistem Rekomendasi Berdasarkan Personalisasi dan Profil Preferensi Bacaan | id |
| dc.title.alternative | Comic Website Development with Recommendation System Based on Personalization and Reading Preference Profile | |
| dc.type | Tugas Akhir | |
| dc.subject.keyword | comic website | id |
| dc.subject.keyword | gamification | id |
| dc.subject.keyword | personalization | id |
| dc.subject.keyword | recommendation system | id |
| dc.subject.keyword | reading preference profile | id |
| dc.subtype | Undergraduate Theses | |