Article Per Year (5 Year)
p-Index From 2021 - 2026
0.778
P-Index
This Author published in this journals
All Journal Jurnal Media Informasi Teknologi Jurnal Nasional Teknologi Informasi dan Aplikasinya
Ida Bagus Gede Sarasvananda
Universitas Udayana
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Optimasi Hyperparameter CART Menggunakan Particle Swarm Optimization (PSO) untuk Klasifikasi Penyakit Stroke I Putu Agus Wahyu Wirakusuma Putra; I Putu Gede Hendra Suputra; Ida Bagus Gede Sarasvananda
Jurnal Nasional Teknologi Informasi dan Aplikasinya Vol. 4 No. 1 (2025): JNATIA Vol. 4, No. 1, November 2025
Publisher : Informatics Department, Faculty of Mathematics and Natural Sciences, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/JNATIA.2025.v04.i01.p18

Abstract

Stroke is a leading cause of death and disability worldwide, including in Indonesia, making early diagnosis crucial. This study aims to enhance the accuracy of stroke classification using the Classification and Regression Tree (CART) algorithm optimized with Particle Swarm Optimization (PSO). A primary challenge in stroke classification is the prevalence of imbalanced datasets. To address this issue, the hybrid sampling technique SMOTEENN (Synthetic Minority Over-sampling Technique-Edited Nearest Neighbors) was applied to balance the class distribution. The standard CART model (baseline) was first evaluated, achieving an accuracy of 94.41%. Subsequently, PSO was implemented to find the optimal hyperparameter combination for the CART model. The PSO optimization results improved the model's performance; the optimized CART model achieved an accuracy of 94.84%, an increase of 0.43% compared to the baseline model. This improvement demonstrates that the combination of the SMOTEENN method for handling imbalanced data and PSO for hyperparameter optimization is an effective and promising approach to enhance the accuracy of stroke classification models.
Segmentasi Pelanggan Berbasis RFMT Menggunakan K-Means dan Hierarchical Clustering I Komang Yosua Triantara; Made Agung Raharja; Ida Bagus Gede Sarasvananda
Jurnal Nasional Teknologi Informasi dan Aplikasinya Vol. 3 No. 4 (2025): JNATIA Vol. 3, No. 4, Agustus 2025
Publisher : Informatics Department, Faculty of Mathematics and Natural Sciences, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/JNATIA.2025.v03.i04.p22

Abstract

The rapid growth of online retail has generated vast transactional data, creating significant opportunities for advanced customer segmentation. While the standard RFM (Recency, Frequency, Monetary) model is widely used in Customer Relationship Management (CRM), it possesses a key limitation by not capturing the temporal dynamics between customer purchases. This research addresses that gap by proposing an RFM-T model, which enhances the traditional framework with Interpurchase Time (IPT) to provide a more holistic view of customer behavior. Using a dual-clustering methodology on an online retail dataset, the K-Means algorithm is first applied for broad segmentation, followed by Hierarchical Clustering to explore deeper sub-segments within high-value groups. The process yielded four primary clusters, and the model's robustness was systematically validated through a strong Silhouette Score, a low Davies-Bouldin Index, and a high Calinski-Harabasz Index. This detailed analysis successfully identified distinct customer personas, such as 'Consistent Loyalists' (low IPT) and 'Periodic Premium Buyers' (high monetary value), which are crucial for developing targeted strategies. The findings demonstrate that this integrated RFM-T framework provides a quantitatively validated with Silhouette Score 0.410, Davies-Bouldin Index 0.720, and Calinski-Harabasz Index 1365.14 this score show actionable model for personalized marketing and effective customer retention.
Klasifikasi Jenis Tari Bali Menggunakan Hyperparameter Tuning CNN dan Transfer Learning ResNet18 I Gede Surya Diva Ananda; Ida Ayu Gde Suwiprabayanti Putra; Ida Bagus Gede Sarasvananda
Jurnal Nasional Teknologi Informasi dan Aplikasinya Vol. 4 No. 3 (2026): JNATIA Vol. 4, No. 3, Mei 2026
Publisher : Informatics Department, Faculty of Mathematics and Natural Sciences, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/JNATIA.2026.v04.i03.p16

Abstract

Balinese dance is a cultural heritage that carries deep philosophical and historical values. In the field of computer vision, image classification of Balinese dance poses a unique challenge due to similarities in movement patterns, costumes, and backgrounds. This research compares two approaches to Balinese dance image classification: a Convolutional Neural Network (CNN) model enhanced with hyperparameter tuning via grid search, and a transfer learning model based on ResNet18. The dataset consists of seven dance classes, each with approximately 240 to 254 images, which are balanced to ensure fair evaluation. The CNN model's hyperparameters, including learning rate, dropout rate, batch size, and optimizer, were optimized using grid search, achieving a top training accuracy of 96.51% and validation accuracy of 72.30%. Meanwhile, the ResNet18 model, leveraging transfer learning from ImageNet, outperformed with a training accuracy of perfect 100% and a validation accuracy of 96.79%. The experimental results suggest that transfer learning significantly boosts performance compared to CNNs trained from scratch, even when carefully tuned. These findings highlight the practical advantage of leveraging pre-trained models in cultural heritage preservation tasks through computer vision.
Implementation of Failover in a Master-Slave Architecture for Soil Moisture Sensor Data Transmission Ajeng Fitria; Uray Ristian; Syamsul Bahri; Ida Bagus Gede Sarasvananda
Jurnal Media Informasi Teknologi Vol. 3 No. 1 (2026): Februari 2026
Publisher : Digital Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69616/mit.v3i1.262

Abstract

Development of Internet of Things (IoT) offers significant opportunities in the agricultural sector by enabling automatic and efficient monitoring of crop conditions. A common issue in IoT-based monitoring systems utilizing a master-slave architecture is the loss of data communication when the master node experiences a failure, which reduces system reliability. This research develops a plant monitoring system based on a master-slave architecture with a failover method using ESP32 and HTTP communication. The system consists of three master nodes, each handling two slave nodes, with a total of twelve soil moisture sensors. Data from the sensors are transmitted by the slave nodes to the master nodes, then forwarded to a cloud server. The failover method functions to maintain communication continuity by rerouting data transmission when a master node experiences a failure. The test results show that the system without failover experienced data loss and communication downtime equal to the duration of the master node failure, while the integration of IoT and failover method maintained communication with an average packet loss of 2.22% and a switching delay of approximately 0.78 seconds.
Co-Authors Ajeng Fitria I Gede Surya Diva Ananda I Komang Yosua Triantara I Putu Agus Wahyu Wirakusuma Putra I Putu Gede Hendra Suputra Ida Ayu Gde Suwiprabayanti Putra Raharja, Made Agung Syamsul Bahri Uray Ristian