Article Per Year (5 Year)
p-Index From 2021 - 2026
1.168
P-Index
This Author published in this journals
All Journal Techno.Com: Jurnal Teknologi Informasi Sinkron : Jurnal dan Penelitian Teknik Informatika Jambura Journal of Informatics Building of Informatics, Technology and Science Journal of Artificial Intelligence and Technology Information Jurnal Informatika dan Rekayasa Perangkat Lunak Journal of Applied Informatics Science
Arnilia, Okma
Universitas Islam Negeri Siber Syekh Nurjati Cirebon
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Engineering

Published : 7 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 7 Documents
Search

 1 
Analisis Sentimen terhadap Pemerintahan Prabowo–Gibran menggunakan IndoBERT dan LDA Ishak, Sahrial Ihsani; Arnilia, Okma; Widodo, Tri; Tatwa, I Gusti Nyoman Agung Bisma
Jambura Journal of Informatics VOL 7, N0 2: OKTOBER 2025
Publisher : Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jji.v1i2.34895

Abstract

This study analyzes public perception of the Prabowo–Gibran administration through online news coverage using a Natural Language Processing (NLP) approach. Data were collected from credible news portals such as Indonesia News and Detik, totaling 195 articles. The analysis was conducted in two stages: first, IndoBERT was used to classify the sentiment into positive, negative, and neutral; second, Latent Dirichlet Allocation (LDA) was applied to identify the main topics driving rage. Sentiment results showed that most topics, particularly those related to the economy, public policy, and governance, were dominated by negative sentiment (80%), while positive sentiment accounted for 15.9% and neutral sentiment for 4.1%. These findings indicate public criticism and concern regarding the effectiveness of policies and economic stability. The combined IndoBERT and LDA approach proved effective in providing a comprehensive understanding of public opinion dynamics in the digital era. It can serve as a consideration for the government in formulating more responsive and transparent communication strategies.Penelitian ini menganalisis persepsi publik terhadap kepemimpinan Prabowo–Gibran melalui pemberitaan media online menggunakan pendekatan Natural Language Processing (NLP). Data dikumpulkan dari portal berita kredibel seperti Antara News dan Detik dengan total 195 artikel. Analisis dilakukan dalam dua tahap: pertama, IndoBERT digunakan untuk mengklasifikasikan sentimen berita menjadi positif, negatif, dan netral; kedua, Latent Dirichlet Allocation (LDA) diterapkan untuk mengidentifikasi topik utama yang mendominasi pemberitaan. Hasil sentimen menunjukkan bahwa sebagian besar topik, terutama terkait ekonomi, kebijakan publik, dan pemerintahan, didominasi oleh sentimen negatif (80%), sedangkan sentimen positif tercatat 15,9% dan netral 4,1%. Temuan ini mengindikasikan adanya kritik dan keprihatinan publik terhadap efektivitas kebijakan dan stabilitas ekonomi. Hasil menunjukkan bahwa sebagian besar topik, terutama terkait ekonomi, kebijakan publik, dan pemerintahan, didominasi oleh sentimen negatif. Temuan ini mengindikasikan adanya kritik dan keprihatinan publik terhadap efektivitas kebijakan dan stabilitas ekonomi. Pendekatan kombinatif IndoBERT dan LDA terbukti efektif dalam memberikan pemahaman komprehensif mengenai dinamika opini publik di era digital, serta dapat menjadi bahan pertimbangan bagi pemerintah dalam merumuskan strategi komunikasi yang lebih responsif dan transparan.
Klasifikasi Risiko Bencana di Indonesia Menggunakan SVM dan Random Forest Erland Adhe Sharendra; Tri Widodo; Damayanti Damayanti; Okma Arnilia
Building of Informatics, Technology and Science (BITS) Vol 8 No 1 (2026): June 2026
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v8i1.9818

Abstract

Indonesia is a country with a high level of disaster vulnerability, requiring effective methods to accurately classify disaster risk levels. This study aims to analyze and compare the performance of Support Vector Machine (SVM) and Random Forest algorithms in disaster risk classification. The dataset used consists of disaster event data from 2019–2024, including disaster type, region, number of victims, and population density. Disaster risk levels were classified into three categories, namely low, medium, and high, based on the total impact calculated from the number of victims. The proposed method includes data preprocessing, normalization, and train-test data splitting. The results show that both models achieved high performance, where Random Forest obtained an accuracy of 95.66% and SVM achieved 95.28%, with ROC-AUC values of 0.9823 and 0.9769, respectively. Random Forest demonstrated slightly better performance with an accuracy difference of 0.38% and more consistent prediction results. The high performance indicates that the models were able to recognize the main patterns within the dataset, although the results were also influenced by the characteristics of the data used. Overall, Random Forest is more suitable for disaster risk classification on data with complex characteristics.
Comparative Analysis of CatBoost and LightGBM for Tree Seedling Survival Prediction to Support Smart Forestry Angga Bayu Santoso; Okma Arnilia; Sahrial Ihsani Ishak; I Gusti Nyoman Agung Bisma Tatwa
Techno.Com Vol. 25 No. 2 (2026): May 2026
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v25i2.15989

Abstract

Tree seedling survival is a critical factor in forest regeneration and sustainable ecosystem management. However, predicting seedling survival remains challenging due to complex interactions between environmental conditions, soil biotic factors, and functional plant traits. This study aims to compare the performance of CatBoost and Light Gradient Boosting Machine (LightGBM) algorithms in predicting tree seedling survival using a machine learning approach. The dataset, obtained from the Tree Survival Prediction dataset on Kaggle, includes environmental variables, soil interaction factors, and functional traits. The target variable is binary, indicating whether a seedling survives or not. Data preprocessing involved handling missing values, encoding categorical variables, normalization, and model validation using 10-fold cross-validation. Model performance was evaluated using accuracy, precision, recall, F1-score, and Receiver Operating Characteristic Area Under Curve (ROC-AUC). The results show that LightGBM outperforms CatBoost, achieving an accuracy of 0.8456, precision of 0.8718, recall of 0.8553, F1-score of 0.8635, and ROC-AUC of 0.9282. In comparison, CatBoost achieves an accuracy of 0.8223 and ROC-AUC of 0.9132. Feature importance analysis indicates that arbuscular mycorrhizal fungi, phenolics, and lignin are the most influential factors affecting seedling survival. These findings demonstrate that LightGBM is a reliable and efficient model for smart forestry applications, supporting data-driven decision-making and improving reforestation strategies. The model enables simulation of planting scenarios, improving resource efficiency and restoration success rates. Keywords - CatBoost, LightGBM, Machine Learning, Seedling Survival, Smart Forestry
Usability Evaluation of a Desktop Library Application Using the System Usability Scale Kahfi Gunardi; layli Hardiyanti; Abdun Wijaya; Okma Arnilia; Fadhlan Ridhwana Sujana
Journal of Applied Informatics Science Volume 1 Issue 2 (2025)
Publisher : GWS Tech Solution

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65897/jais.v1.i2.71

Abstract

This study aims to evaluate the usability of an offline desktop library application developed using Java Swing and integrated with an iReport feature. The evaluation employed the System Usability Scale (SUS) using a descriptive quantitative approach. Data were collected from 28 respondents consisting of librarians and library administrators as the primary system users. The results show that the application achieved an average SUS score of 62.77, which falls into the marginal acceptability range, corresponding to Grade D and an “OK” adjective rating, and below the standard SUS benchmark score of 68. These findings indicate that the application adequately supports basic library operations but still exhibits usability limitations, particularly in interface consistency, user confusion, and reliance on technical assistance. Consequently, improvements in interface design and user experience are necessary to enhance the overall usability of the application.
Clustering of Provincial Health Vulnerability Levels in Indonesia Using the K-Means Method Okma Arnilia; Sahrial Ihsani Ishak; Tri Widodo; I Gusti Nyoman Agung Bisma Tatwa
Jurnal Informatika dan Rekayasa Perangkat Lunak Vol. 7 No. 1 (2026): Volume 7 Number 1 March 2026
Publisher : Universitas Teknokrat Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33365/jatika.v7i1.1469

Abstract

This study aims to classify the health vulnerability levels of 38 provinces in Indonesia based on health and socio-economic indicators in 2024, including the number of hospitals, access to adequate sanitation, access to safe drinking water, stunting prevalence, number of health facilities, population size, and the percentage of poor population. The analysis began with data normalization using the z-score method to standardize variable scales and prevent dominance by indicators with larger value ranges. Following normalization, the optimal number of clusters was determined using the Elbow method by examining the decrease in inertia across different k-values. Based on the inertia pattern and cluster stability, the optimal number of clusters was identified as K=4, which adequately represents the variation in health vulnerability. The clustering results were subsequently visualized in a spatial map using Indonesia’s provincial administrative boundaries. The visualization revealed clear geographical variation across regions, with Cluster 1 representing provinces with very good health conditions, Cluster 2 good conditions, Cluster 3 moderate conditions, and Cluster 4 provinces requiring special attention regarding health indicators. These findings provide a comprehensive overview of health vulnerability distribution in Indonesia and are expected to inform policymakers and stakeholders in prioritizing region-based health interventions, strengthening health development strategies, and promoting more equitable national health services.
Image-Based Food Classification for Nutritional Information Estimation Using Deep Learning Sahrial Ihsani Ishak; Sri Dianing Asri; Bias Yulisa Geni; Okma Arnilia; Tri Widodo; Diva Maulana Ilham
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16170

Abstract

This study aims to develop an image-based food classification application integrated with nutritional information retrieval using a deep learning approach. The proposed system is designed to recognize food types from images and provide nutritional information based on an Indonesian food nutrition database. The method involves collecting a dataset of 8,248 images representing 38 categories of Indonesian traditional foods, performing image preprocessing and data augmentation, and developing a Convolutional Neural Network (CNN) model based on the MobileNetV2 architecture through transfer learning. Model performance was evaluated using a 3-fold stratified cross-validation strategy and measured using accuracy, precision, recall, and F1-score metrics. Experimental results showed that the proposed model achieved average accuracy, precision, recall, and F1-score values of 98.85%, 98.88%, 98.85%, and 98.85%, respectively, demonstrating robust and consistent classification performance across the validation folds. The trained model was subsequently deployed into a mobile application using TensorFlow Lite to support real-time food classification and nutritional information presentation. The main contribution of this study is the development of an end-to-end mobile system that integrates deep learning-based food classification with an Indonesian food nutrition database, enabling users to obtain calorie, protein, fat, and carbohydrate information quickly and conveniently for dietary monitoring and health awareness.  
A Hybrid SITDE Weighting and HYBSO Method in Decision Support Systems for Warehouse Employee Selection Tri Widodo; Okma Arnilia; Iryanto Chandra; Sahrial Ihsani Ishak; Setiawansyah Setiawansyah
Journal of Artificial Intelligence and Technology Information (JAITI) Vol. 4 No. 3 (2026): Volume 4 Number 3 September 2026
Publisher : PT. Tech Cart Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58602/jaiti.v4i3.344

Abstract

Warehouse employee selection is a complex process because it involves various heterogeneous criteria, both quantitative and qualitative, such as physical tests, accuracy, work experience, discipline, and teamwork ability. A selection process that still relies on subjective judgment risks producing inconsistent and suboptimal decisions. This study proposes a hybrid method that integrates skewness impact through distributional evaluation (SITDE) for objective determination of criteria weights with a hybrid solution algorithm (HYBSO) for systematic ranking of alternatives. The results of applying this method indicate that Work Experience has the highest weight of 0.2813, followed by the Discipline Test (0.1969) and Physical Test (0.1926), while the Accuracy Test (0.1641) and Teamwork Test (0.1652) have lower weights. The final ranking results show that Candidate HW ranks first with a score of 0.3596, followed by Candidate FN (0.7066) and Candidate DK (0.9162). The sensitivity analysis evaluates the stability of the ranking results under changes in criterion weights across 20 scenarios, demonstrating the robustness of the proposed SITDE-HYBSO method and identifying candidates whose ranking positions are more sensitive to variations in criterion importance. These findings confirm that the SITDE-HYBSO hybrid method is capable of producing objective, consistent, and accountable employee selection decisions, while also providing an adaptive mechanism to respond to changes in organizational priorities.
Co-Authors Abdun Wijaya Angga Bayu Santoso Bias Yulisa Geni Damayanti, Damayanti Diva Maulana Ilham Erland Adhe Sharendra Fadhlan Ridhwana Sujana I Gusti Nyoman Agung Bisma Tatwa I Gusti Nyoman Agung Bisma Tatwa Iryanto Chandra Ishak, Sahrial Ihsani Kahfi Gunardi layli Hardiyanti Sahrial Ihsani Ishak Sahrial Ihsani Ishak Setiawansyah Setiawansyah Sri Dianing Asri Tatwa, I Gusti Nyoman Agung Bisma Tri Widodo Tri Widodo Tri Widodo