Claim Missing Document
Check
Articles

Found 4 Documents
Search

Sentiment Analysis of Public Opinions Regarding "Ideas of Presidential Candidates" in YouTube Video Comments with Robustly Optimized BERT Pretraining Approach Sumihar, Yoel Pieter
Jurnal Sistem Informasi dan Ilmu Komputer Vol. 8 No. 1 (2024): JUSIKOM: JURNAL SISTEM INFROMASI ILMU KOMPUTER
Publisher : Fakultas Teknologi dan Ilmu Komputer Universitas Prima Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34012/jurnalsisteminformasidanilmukomputer.v8i1.5350

Abstract

Social media and video-sharing platforms such as YouTube have become one of the primary sources of information and social interaction in modern society. In politics, YouTube has become essential for spreading ideas, campaign platforms, and opinions about the presidential election. Using the pre-trained Indonesian Roberta Base Sentiment Classifier Model, the data obtained from YouTube comments will be divided into three labels: positive, negative, and neutral. The results of this study are the accuracy for each sentiment label, where the value for positive is 93%, the negative is 90.5%, and the neutral is 93.04%. Residents give more positive comments to presidential candidate Prabowo Subianto, with a positive value of 54.13%, followed by Anies Baswedan at 42.8% and Ganjar Pranowo at 31.91%.
Implementation of Microservices Architecture in a Retail Web Application Using Apache Kafka as a Message Broker Daeli, Stefanus; Lase, Kristian Juri Damai; Sumihar, Yoel Pieter
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 2 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i2.13932

Abstract

Web-based applications are often initially developed using monolithic architecture due to its simplicity and ease of deployment. However, as application complexity grows, monolithic systems face critical limitations in scalability, flexibility, and performance. This research applies a microservices architecture to a Retail Web divided into four core services: user, product, transaction, and notification management. Apache Kafka is integrated as a message broker to support asynchronous, real-time communication across services. A total of 2,001 requests were recorded during system testing using Prometheus. The srv_tulityretailaccounts service achieved an average response time of 122.8 ms, and the srv_tulityretailtransactions service maintained 188.1 ms with a 98% success rate. The srv_tulityretailproducts service also demonstrated stable performance with consistently low response times and no error spikes. Meanwhile, the srv_tulityretailnotifications service showed the highest efficiency with an average response time of 28.5 ms, CPU usage at 12.75% (1.53 of 12 cores), and memory usage at 2.07 GB (56.5%) of 3.66 GB. Throughout testing, no service exhibited resource saturation or degradation, even under concurrent load conditions. This confirms the system’s horizontal scalability, where each service can independently scale without impacting others. Overall, the microservices approach has proven effective in enhancing performance, modularity, and production-readiness, while laying a strong foundation for continuous integration, deployment automation, and future feature expansion.
E-Library for Blankspot Areas at SMP Negeri Satap 1 Kolaka Utara Sumihar, Yoel Pieter; Adinda, Adinda; Bantun, Suharsono; Jum'ah, Muhammad Na'im Al Jum'ah
Jurnal Media Informasi Teknologi Vol. 1 No. 1 (2024): Februari 2024
Publisher : Digital Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69616/mit.v1i1.164

Abstract

Digital libraries (digital libraries) or also known as electronic libraries (e-libraries) are introduced with the advantage of providing collections in electronic form that can be accessed by users without limitations of place and time by using electronic media such as computers, laptops or gadgets connected to computers. The E-Library information system for areas without a signal at SMPN Satap 1 Kolaka Utara is a data processing system that is enabled to process new member registration data, visit data processing, transaction data processing, book data processing, and fine processing. The research method uses the method of literature study, observation, interviews, and the system development method uses the waterfall. The analysis and design tools used are data flow diagrams (DFD) and database design using Entity Relationship Diagrams (ERD), the programming language used is hypertext preprocessor (PHP), and the database uses MySQL. The test method uses Blackbox and UAT testing. From the results of the discussion in the previous chapters, it can be concluded that the E-Library information system for areas without a signal at SMPN Satap 1 Kolaka makes it easy for members to read books offline and makes it easier to manage book data and borrow books.
Analisis Performa Metode KNN, Yolov8, Dan Yolov11 Pada Klasifikasi Konjungtiva Mata Untuk Deteksi Anemia Sumihar, Yoel Pieter; Maedjaja, Febe; Sas, Valentino Henry
TIN: Terapan Informatika Nusantara Vol 6 No 7 (2025): December 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/tin.v6i7.8707

Abstract

Rapid and non-invasive anemia detection is crucial, especially in regions with limited laboratory facilities. The conjunctiva of the eye serves as a promising visual indicator for anemia through the analysis of color and texture. This study aims to analyze and compare the performance of three image classification methods K-Nearest Neighbors (KNN), YOLOv8, and YOLOv11 in detecting anemia using conjunctival images. The CP-AnemiC dataset was employed, consisting of 710 original images, later expanded to 3,550 images through augmentation. KNN utilized color features extracted from the CIE LAB color space, while YOLOv8 and YOLOv11 leveraged automatic feature extraction using convolutional neural networks. Evaluation metrics included accuracy, precision, recall, and F1-score. The results indicate that YOLOv8 achieved the best performance with 93.4% accuracy and a 94.5% F1-score, followed by YOLOv11 with 93.0% accuracy and a 94.2% F1-score. In contrast, KNN obtained an accuracy of only 85.7%. YOLOv8 demonstrated fast and accurate detection, while YOLOv11 exhibited more stable training behavior. These findings highlight that deep learning models particularly YOLOv8 and YOLOv11 are highly promising for implementing efficient, accurate, and practical conjunctival image–based anemia detection systems. This research contributes by presenting an explicit comparative analysis between the classical method (KNN) and the latest deep learning models (YOLOv8 and YOLOv11) in the specific context of conjunctival image classification.