Article Per Year (5 Year)
p-Index From 2021 - 2026
1.376
P-Index
This Author published in this journals
All Journal Teika Syntax Idea Jurnal Teknologi Informasi : Jurnal Keilmuan dan Aplikasi Bidang Teknik Informatika Community Engagement and Emergence Journal (CEEJ) Jurnal Teknoif Teknik Informatika Institut Teknologi Padang Jurnal Pengabdian Masyarakat Bidang Sains dan Teknologi Jurnal Ilmiah Sistem Informasi
Yusran Timur Samuel
Unknown Affiliation
Aerospace Engineering Agriculture, Biological Sciences & Forestry Humanities Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Dentistry Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Health Professions Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Library & Information Science Mathematics Mechanical Engineering Medicine & Pharmacology Nursing Public Health Transportation Veterinary Other

Published : 26 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 26 Documents
Search

 1   2   3 
Analisis Pengaruh Nilai Evaluasi Dosen Terhadap Kelulusan Mata Kuliah Mahasiswa Universitas Advent Indonesia dengan Decision Tree Yusran Tarihoran; Timotius Ginting
Syntax Idea 1131-1137
Publisher : Ridwan Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46799/syntax-idea.v6i3.3088

Abstract

Analysis of course completion at Universitas Advent Indonesia (UNAI) is an important thing to do. By knowing the course completion of students early, UNAI can take necessary actions to improve students' course completion. This study reveals the influence of lecturer evaluation scores on student course completion in the Web Programming 1, Web Programming 2, and Programming Algorithm 1 courses during the academic years 2020-2021 to 2022-2023. The purpose of this study is to see the influence of lecturer evaluation scores on student course completion in these courses and which competencies most affect student scores. The method used in this study is data mining with the C4.5 decision tree algorithm. The results of this study show that out of 5 competencies tested, 3 competencies affect student course completion. This study uses a confusion matrix evaluation model that produces a prediction accuracy of 80%.
Implementasi Sistem Internet of Things (IoT) pada PT XXX menggunakan sistem MQTT Sumartono, Katherine Febrianty; Samuel, Yusran Timur
TeIKa Vol 14 No 2 (2024): TeIKa: Oktober 2024
Publisher : Fakultas Teknologi Informasi - Universitas Advent Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36342/teika.v14i2.3784

Abstract

The Internet of Things (IoT) system has been implemented at PT XXX to improve operational efficiency, especially in monitoring the Ingersoll C1000 centrifugal compressor. This compressor produces high-pressure air that plays an important role in the nickel processing process. This system integrates the Message Queuing Telemetry Transport (MQTT) protocol, IoT sensors, Teltonika RUT951 gateway, and Mosquitto broker to send temperature data in JSON format in real-time. This data is then visualized using AVEVA PI Vision, which provides trend-based information to support analysis and decision making. Previously, monitoring was only carried out locally without historical data, making it difficult to detect potential damage. Operators also had to conduct direct inspections to the field to check the condition of the compressor, which was time-consuming and inefficient. With this IoT-based system, data can be monitored directly from the head office, saving travel time and increasing operational productivity. Trials in a real operational environment showed that data transmission via MQTT was stable even in a limited network. Visualization in AVEVA PI Vision provides a real-time picture that supports early detection of anomalies and data-driven decision making. The results of this study indicate that IoT technology with the MQTT protocol is able to improve monitoring efficiency, optimize system performance, and provide innovative solutions to connectivity challenges in the industry. This implementation also opens up new opportunities for IoT applications in supporting operational efficiency and sustainability.
IDENTIFIKASI KEPADATAN PENDUDUK DI PROVINSI JAWA BARAT MENGGUNAKAN HIERARCHICAL CLUSTERING Simanjuntak, Wahyu Iskandar; Samuel, Yusran Timur
Jurnal Teknoif Teknik Informatika Institut Teknologi Padang Vol 13 No 1 (2025): TEKNOIF APRIL 2025
Publisher : ITP Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21063/jtif.2025.V13.1.28-39

Abstract

This research applies a hierarchical clustering algorithm to identify population density patterns in West Java (18 regencies, 9 cities) as a basis for natural disaster management. Population density data for 2020-2022 from the West Java Population Office was analyzed to group areas into three categories: densest, medium, and lowest. The hierarchical clustering method was used to group areas based on population density and flood potential, with the additional attribute of river presence. The clustering results were evaluated using the Davies-Bouldin index. The results showed that the algorithm was successfully applied, grouping 20 districts/cities with the lowest population density (Cluster 0), 3 districts/cities with medium density (Cluster 1), and 4 districts/cities with the densest density (Cluster 2).This research is expected to provide insight to the government and related institutions in planning disaster mitigation based on population density patterns, so as to reduce the risk of natural disasters in the future. This research takes data from the official source https://jabar.bps.go.id/indicator/12/245/1/kepadatan-penduduk-menurut-kabupaten-kota.html. The main objective of this research is to understand population density patterns that can provide an indication of the level of risk to certain natural disasters in the West Java region. This information is expected to be used as a basis for more effective and efficient disaster management strategies in the future. The implication of this research shows that by understanding the pattern of population density and river distribution through the hierarchical clustering method, the government and related institutions can formulate more targeted disaster management strategies.
Peningkatan Pemahaman Siswa terhadap Proses Pembuatan Game Edukatif di Sekolah Perguruan Advent Parongpong Henry Pandia; Sintaria Sembiring; Yusran Timur Samuel
ABDIKAN: Jurnal Pengabdian Masyarakat Bidang Sains dan Teknologi Vol. 4 No. 2 (2025): Mei 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/abdikan.v4i2.5409

Abstract

This community service program aimed to enhance digital literacy and student creativity among junior high school students through training in the development of educational games using the Scratch platform. The training was conducted in two sessions at Perguruan Advent Parongpong School and involved 8th and 9th grade students. The first session introduced basic visual programming concepts through a simple project titled Underwater Life, while the second session guided students to create a game called Bat vs Dragonfly, featuring a point system and win-lose logic. The method used was active and participatory, including material presentations, demonstrations, and hands-on practice with assistance. Questionnaire results revealed that most students regularly play games, showed strong interest in learning to create games, and viewed the use of games in learning positively. In the first session, 8 out of 29 students successfully completed the project, while in the second session, 21 out of 30 students completed the game within the allotted time. This activity demonstrated that educational games can effectively foster student interest in learning, logical thinking, and creativity. The outcomes are expected to serve as a foundation for schools to develop extracurricular programs focused on creative technology, and to encourage parents and educators to guide the use of technology in a positive and productive direction.
Analisis Sentimen Isu Megathrust Indonesia Di Twitter Menggunakan Support Vector Machine Dan Naive Bayes Christanto, Arya; Timur Samuel, Yusran
Jurnal Ilmiah Sistem Informasi (JISI) Vol. 4 No. 1 (2025): MARET
Publisher : Universitas Muhammadiyah Metro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24127/jisi.v4i1.8453

Abstract

Indonesia berisiko tinggi mengalami gempa megathrust yang berpotensi menimbulkan bencana besar, termasuk tsunami. Ancaman ini mendapat perhatian karena dapat merusak infrastruktur, mengganggu komunikasi, dan berdampak signifikan terhadap perekonomian. Media sosial, terutama Twitter, merupakan wadah setiap individu atau masyarakat dalam bertukar informasi, menyampaikan ceramah, dan memberikan pendapat terkait isu ini. Penelitian ini menganalisis sentimen publik di Twitter mengenai megathrust di Indonesia dengan metode algoritma Support Vector Machine (SVM) serta Naive Bayes (NVB). Data dikumpulkan dari 404 tweet berbahasa Indonesia yang diposting antara 1 Agustus hingga 30 September 2024. Setelah melalui pra-pemrosesan, data yang diperoleh dilabeling secara manual dan otomatis sebelum diklasifikasikan dengan RapidMiner. Hasil pada penelitian ini menjelaskan bahwa Naive Bayes memiliki tingkat akurasi yang lebih tinggi (83,58%) dibandingkan SVM (75,94%). Selain itu, NVB lebih unggul dalam mengenali sentimen negatif dengan recall sebesar 68%. Analisis tersebut memberikan dampak terbaru, terutama wawasan mengenai persepsi individua tau masyarakat terhadap megathrust dan dapat menjadi dasar dalam merancang strategi komunikasi yang lebih efektif. Dengan memahami respons masyarakat, pihak yang berwenang dapat menyusun kebijakan mitigasi bencana yang lebih tepat dan meningkatkan kesiapsiagaan masyarakat. Penelitian ini juga menyoroti pentingnya media sosial sebagai sumber data dalam kajian kebencanaan, khususnya dalam memahami reaksi dan kesiapan masyarakat terhadap ancaman gempa bumi
Perbandingan Kinerja Arsitektur LeNet-5 dan AlexNet dalam Klasifikasi Daun Teh Siap Panen Menggunakan Convolutional Neural Network Samuel, Yusran Timur; Rundengan, Juwita
Community Engagement and Emergence Journal (CEEJ) Vol. 7 No. 3 (2026): Community Engagement & Emergence Journal (CEEJ)
Publisher : Yayasan Riset dan Pengembangan Intelektual

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37385/ceej.v7i3.10501

Abstract

Industri teh memiliki peran penting dalam sektor agribisnis Indonesia, dengan kualitas teh yang sangat dipengaruhi oleh pemilihan daun yang tepat dan waktu pemetikan yang sesuai. Klasifikasi daun teh masih dilakukan secara manual, yang mengandalkan pengalaman petani dan dapat dipengaruhi oleh faktor eksternal seperti kelelahan. Penelitian ini bertujuan untuk membandingkan kinerja dua arsitektur Convolutional Neural Network (CNN), yaitu LeNet-5 dan AlexNet, dalam mengklasifikasikan daun teh siap panen dan belum siap panen berdasarkan metrik akurasi, presisi, recall, F1-score, loss, serta efisiensi komputasi.Dataset yang digunakan dalam penelitian ini terdiri dari 4.198 gambar daun teh yang telah dilabeli. Model LeNet-5 dan AlexNet dilatih dan diuji menggunakan dataset ini, dan hasil evaluasi menunjukkan perbedaan performa yang signifikan antara keduanya. Model AlexNet menghasilkan akurasi 83,33%, dengan precision 0,86, recall 0,90, dan F1-score 0,88 untuk kelas "Belum Siap Panen", serta precision 0,76, recall 0,69, dan F1-score 0,72 untuk kelas "Siap Panen". Sementara itu, model LeNet-5 mencapai akurasi 71,67%, dengan precision 0,75, recall 0,88, dan F1-score 0,81 untuk kelas "Belum Siap Panen", namun hanya memperoleh precision 0,59, recall 0,36, dan F1-score 0,45 untuk kelas "Siap Panen". Meski LeNet-5 lebih efisien dalam hal waktu pemrosesan, AlexNet lebih unggul dalam akurasi dan kemampuan menangkap pola yang lebih kompleks pada data daun teh. Kedua model menunjukkan gejala overfitting, tetapi secara keseluruhan AlexNet lebih efektif untuk klasifikasi daun teh siap panen. Penelitian ini menyimpulkan bahwa AlexNet lebih optimal dalam mengklasifikasikan daun teh dengan akurasi dan kinerja yang lebih baik dibandingkan dengan LeNet-5, meskipun membutuhkan sumber daya komputasi yang lebih besar.
Co-Authors Abdul Fatah Nordin Adrian R. Lemuel Antonius Maryo Manurung Bern Jonathan Boston Pahala Siahaan Choresy Michael G. Butar-butar Christanto, Arya Chrystle Beatrix Allbright Nahuway Elmor Benedict Wagiu Frengky Simbolon Heima Sitorus Henry Pandia Hiras Parasian Doloksaribu Ira Worang Josephine Naibaho Kemala DEwi Kevin Jeremy Manurip Marlinda Siahaan Raja Simarmata Raymond Maulany Riza Muhida Rundengan, Juwita Simanjuntak, Wahyu Iskandar Sintaria Sembiring Suhaimi B Mohd Zaid Sumartono, Katherine Febrianty Timotius Ginting Yank Nekmese K.