Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Jurnal Sistem Informasi dan Sistem Komputer Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI) Jurnal Ilmu Komputer dan Teknologi (IKOMTI)
Rooy Marthen Thaniket
Universitas Satya Wiyata Mandala
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Other

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Klasifikasi Jenis Burung Cendrawasih Menggunakan Convolutional Neural Network (CNN) Berdasarkan Citra Rooy Marthen Thaniket; Musa Henri Janto Rahanra; Wardhana Wahyu Dharsono
Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI) Vol 8 No 1 (2025): Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat Universitas Lamappapoleonro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57093/jisti.v8i1.280

Abstract

Birds of paradise are iconic symbols of Indonesia's biodiversity, especially in Papua, with more than 40 recorded species. Manual classification requires specific expertise and is time-consuming. This study aims to develop an automated classification system for birds of paradise using Convolutional Neural Network (CNN), specifically the MobileNetV2 architecture known for its efficiency in image processing. The dataset used comprises three species: Cicinnurus regius, Paradisaea apoda, and Paradisaea rubra. The preprocessing steps include image augmentation, resizing, and normalization. The training results show an accuracy of 98.49% and validation accuracy of 97.50%. Evaluation using a confusion matrix reveals high accuracy and minimal misclassification. This model shows great potential for use in conservation applications and automatic bird species identification
Analisis Sentimen Keluhan Layanan Cash on Delivery (COD) pada Platform Shopee Menggunakan IndoBERT Musa Rahanra; Rooy marthen Thaniket; Nicodemus Rahanra; Hermanus J Suripatty
Jurnal IT UHB Vol 7 No 2 (2026): Jurnal Ilmu Komputer dan Teknologi
Publisher : Universitas Harapan Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35960/ikomti.v7i2.2545

Abstract

This study aims to analyze user review sentiment regarding the Cash on Delivery (COD) service on the Shopee e-commerce platform using the IndoBERT model. The dataset consisted of 1,840 unique Google Play Store reviews related to COD collected from September 25, 2018, to December 29, 2023. The stages of the research included data collection, text preprocessing, sentiment labeling, and modeling using IndoBERT. The results show that 48.4% of the reviews were negative, 41.1% positive, and 10.5% neutral. The IndoBERT model achieved an accuracy of 92.1%, with a weighted precision of 0.923, weighted recall of 0.921, and weighted F1-score of 0.922. The results indicate that the IndoBERT method is capable of classifying sentiments with good performance. The findings provide an overview of users’ perceptions of the COD service on Shopee. However, this study is limited to the use of a dataset from a single platform and cannot be generalized to all e-commerce platforms in Indonesia. Future research should use cross-platform datasets to obtain more comprehensive results.
Prediksi Kegagalan Build Real-Time Pada Pipeline CI/CD Menggunakan Pendekatan Machine Learning Berbasis Data Streaming Hirmayanti Hirmayanti; Harianto Harianto; Rooy Marthen Thaniket
Jurnal Sistem Informasi dan Sistem Komputer Vol 11 No 2 (2026): Vol 11 No 2 - 2026
Publisher : STIMIK Bina Bangsa Kendari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51717/simkom.v11i2.1414

Abstract

Kegagalan proses build pada lingkungan Continuous Integration/Continuous Deployment (CI/CD) dapat menghambat rilis perangkat lunak, meningkatkan biaya perbaikan, dan menurunkan efisiensi pengembangan. Hal ini menunjukkan perlunya sistem prediksi yang cepat, adaptif, serta mampu menangani data yang terus mengalir secara real-time. Penelitian ini bertujuan untuk menganalisis dan membandingkan kinerja beberapa algoritma online machine learning dalam memprediksi kegagalan build pada pipeline CI/CD. Dataset yang digunakan berasal dari TravisTorrent yang mencakup aktivitas repositori, perubahan kode, metrik pengujian, serta riwayat build. Tahapan penelitian meliputi preprocessing data, pembentukan skema data streaming, penerapan model Hoeffding Tree, Adaptive Random Forest (ARF), dan Online Logistic Regression, serta evaluasi menggunakan metrik accuracy. Hasil eksperimen menunjukkan model terbaik mencapai accuracy 97,35%, lebih tinggi dibandingkan penelitian sebelumnya sebesar 95,9%. Peningkatan absolut sebesar 1,45 poin persentase atau sekitar 1,51% ini membuktikan bahwa pendekatan yang digunakan lebih efektif dalam mengenali pola kegagalan build pada lingkungan CI/CD berbasis streaming data.
Co-Authors Dharsono, Wardhana Wahyu Harianto Harianto Hermanus J Suripatty Hirmayanti Hirmayanti Musa Henri Janto Rahanra Musa Rahanra Rahanra, Nicodemus