Article Per Year (5 Year)
p-Index From 2021 - 2026
2.162
P-Index
This Author published in this journals
All Journal Seminar Nasional Aplikasi Teknologi Informasi (SNATI) Teknik dan Ilmu Komputer Proceeding SENDI_U Perspektif : Jurnal Ekonomi dan Manajemen Universitas Bina Sarana Informatika Informatics for Educators and Professional : Journal of Informatics Prosiding Seminar Nasional Teknoka IKRA-ITH Informatika : Jurnal Komputer dan Informatika JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI Jurnal Nasional Komputasi dan Teknologi Informasi CCIT (Creative Communication and Innovative Technology) Journal INFOKUM Jurnal Scientia Nusantara Science and Technology Proceedings Jurnal Pustaka Mitra : Pusat Akses Kajian Mengabdi Terhadap Masyarakat Jurnal Pengabdian Mandiri
Agni Isador Harsapranata
Universitas Bina Sarana Informatika
Aerospace Engineering Agriculture, Biological Sciences & Forestry Humanities Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Materials Science & Nanotechnology Mathematics Medicine & Pharmacology Neuroscience Public Health Social Sciences Transportation Other

Published : 22 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 22 Documents
Search

 1   2   3 
Analisis Segmentasi Tayangan Netflix Berdasarkan Metadata Menggunakan PCA, Multi-Model Clustering, dan Validasi Stabilitas Klaster Wahyu Pratama; Agni Isador Harsapranata; Ahmad Rais Ruli

Publisher :

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32672/jnkti.v8i5.9676

Abstract

Abstrak - Pertumbuhan layanan streaming seperti Netflix menghadirkan tantangan dalam memahami preferensi pengguna serta mengelola ribuan konten yang tersedia. Oleh karena itu, diperlukan metode analisis segmentasi untuk mengidentifikasi pola distribusi konten secara lebih efisien. Penelitian ini bertujuan untuk melakukan segmentasi konten Netflix menggunakan pendekatan clustering dengan algoritma K-Means, serta membandingkan performanya dengan metode lain. Dataset yang digunakan terdiri dari 15.026 konten dengan 1.206 fitur, yang direduksi menggunakan Principal Component Analysis (PCA) menjadi 1.134 komponen dengan 95% varians tetap terjaga. Evaluasi kinerja model dilakukan menggunakan Silhouette Score dan Adjusted Rand Index (ARI). Hasil penelitian menunjukkan bahwa K-Means dengan 9 cluster merupakan metode terbaik dengan nilai Silhouette Score 0,1481 dan ARI 0,6381. Segmentasi menghasilkan distribusi yang didominasi oleh Segmen 0 (81,5%) sebagai segmen utama dan Segmen 4 (18,4%) sebagai segmen potensial, sementara segmen lainnya (1%) dikategorikan sebagai segmen niche. Temuan ini memberikan implikasi akademis berupa kontribusi dalam pengembangan metode segmentasi berbasis machine learning, serta implikasi praktis bagi industri streaming untuk mengoptimalkan strategi investasi konten, mengembangkan sistem rekomendasi, dan mengeksplorasi peluang pasar baru.Kata Kunci: Netflix; Klastering; PCA; K-Means; Silhouette Score; Adjusted Rand Index; Abstract - The rapid growth of streaming services such as Netflix poses significant challenges in understanding user preferences and managing the vast number of available contents. Therefore, content segmentation methods are required to efficiently identify distribution patterns. This study aims to perform Netflix content segmentation using the clustering approach with the K-Means algorithm, and to compare its performance with alternative methods. The dataset consists of 15,026 contents with 1,206 features, which were reduced using Principal Component Analysis (PCA) into 1,134 components while preserving 95% variance. Model performance was evaluated using the Silhouette Score and Adjusted Rand Index (ARI). The results indicate that K-Means with 9 clusters achieved the best performance, yielding a Silhouette Score of 0.1481 and an ARI of 0.6381. The segmentation revealed that Segment 0 (81.5%) dominated as the main segment and Segment 4 (18.4%) was identified as a potential segment, while the remaining clusters (1%) represented niche segments. These findings provide both academic implications by contributing to the development of machine learning-based segmentation methods, and practical implications for the streaming industry to optimize content investment strategies, improve recommendation systems, and explore new market opportunities.Keywords: Netflix; Clustering; PCA; K-Means; Silhouette Score; Adjusted Rand Index;
Arsitektur Adaptive Sleep Scheduling Berbasis Harvesting Energy pada Sistem Sensor Debit Air IoT untuk Kawasan Tanpa Sumber Listrik Tetap: Arsitektur Adaptive Sleep Scheduling Berbasis Harvesting Energy pada Sistem Sensor Debit Air IoT untuk Kawasan Tanpa Sumber Listrik Tetap Ahmad Rais Ruli; Agni Isador Harsapranata
INFORMATICS FOR EDUCATORS AND PROFESSIONAL : Journal of Informatics Vol. 11 No. 1 (2026): INFORMATICS FOR EDUCATORS AND PROFESSIONAL : JOURNAL OF INFORMATICS (Juni 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat Universitas Bina Insani

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51211/itbi.v11i1.3895

Abstract

Real-time water flow monitoring in rural areas without permanent electricity access (off-grid) is a major challenge in water resource management. Limited energy in battery-based IoT devices is the main obstacle to operational sustainability. This study proposes an Adaptive Sleep Scheduling Architecture Based on Energy Harvesting for an IoT Water Flow Sensor System. The system uses UNO R4 ESP-32 Wifi and ESP-32 S3 W-ROOM microcontrollers, integrating a solar energy harvesting unit with an adaptive sleep scheduling algorithm that dynamically adjusts sensor active/sleep cycles based on battery State of Charge (SoC) and harvested energy availability. The experimental method covers four stages: (1) hardware design and integration, (2) adaptive algorithm development, (3) off-grid field testing, and (4) comparative evaluation against fixed scheduling. Target outcomes include a minimum 30% battery lifetime improvement, maintained Packet Delivery Ratio under low-energy conditions, and autonomous continuous operation without dependence on grid electricity. This research contributes to water infrastructure resilience in remote areas and supports SDG 6 for sustainable clean water management. Keywords: IoT; Adaptive Sleep Scheduling; Energy Harvesting; Water Flow Sensor; Off-Grid Power Source
Co-Authors Ahmad Rais Ruli Ahmad Rais Ruli Ana Ramadhayanti Andi Martias Danny Manongga Dewi Astuti Joko Ariawan Maksum Maksum Omar Pahlevi Priatno, Priatno Rahma Wiyanti Rahmat Novari Sutarto Wijono Tri Hartati, Tri Wahyu Pratama