cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Teguh Wiyono
Contact Email
indexsasi@apji.org
Phone
+6285700037105
Journal Mail Official
indexsasi@apji.org
Editorial Address
Jalan Watunganten 1 No 1-6, Batursari, Mranggen Kab. Demak Jawa Tengah 59567
Location
Kab. demak,
Jawa tengah
INDONESIA
Global Science: Journal of Information Technology and Computer Science
Published by International Forum of Researchers and Lecturers
ISSN : 31089976     EISSN : 31089968     DOI : 10.70062
Core Subject : Science,
Computer Science & IT
Global Science: Journal of Information Technology and Computer Science; This a journal intended for the publication of scientific articles published by International Forum of Researchers and Lecturers This journal contains studies in the fields of Information Technology and Computer Science, both theoretical and empirical. This journal is published 1 year 4 times (March, June, September and December)
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 39 Documents
 1   2   3   4 
Improving Data Security and Verification in Federated Learning with Homomorphic Encryption: Literature Review: Subtitle Alfina Tiur Mida Sitanggang; Eddy Refianto Eddy; Ferry Muhamad Ramadhan; Frangky
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 2 (2026): June: Global Science: Journal of Information Technology and Computer Science
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i2.200

Abstract

Federated Learning has emerged as a prominent solution for collaboratively training machine learning models without sharing raw data, thereby preserving user privacy in today's digital era. However, threats such as man-in-the-middle attacks, data reconstruction attacks, and model manipulation remain significant challenges for this approach. This literature review explores the integration of Homomorphic Encryption (HE) into Federated Learning to enhance data security and ensure model integrity verification. The findings indicate that the combination of Homomorphic Encryption and Federated Learning can reduce the risk of data leakage by up to 90% compared to conventional non-encrypted methods. Furthermore, despite introducing a computational overhead of approximately 20–30%, model accuracy remains relatively high, with only a 1–2% reduction. This study contributes to the development of a more secure, efficient, and reliable Federated Learning framework for critical applications, including healthcare, finance, and the Internet of Things (IoT). Keywords: Federated Learning, Homomorphic Encryption, Data Security, Model Verification, Privacy Protection.
Explainable AI for Predicting Hypertension Risk in Diabetic Patients Using Model Interpretability Techniques Samsinar; Cyntia Lasmi Andesti; Ummul Fitri Afifah
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 2 (2026): June: Global Science: Journal of Information Technology and Computer Science
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i2.201

Abstract

Hypertension is a major comorbidity in diabetic patients, significantly increasing the risk of cardiovascular diseases and other complications. Despite advancements in machine learning for predicting hypertension, challenges such as class imbalance and lack of model interpretability remain barriers to clinical adoption. This study aims to bridge these gaps by integrating Explainable AI (XAI) techniques with resampling methods to predict hypertension risk in diabetic patients. This study applied SMOTE, SMOTE-ENN, and threshold tuning to improve model performance while maintaining interpretability. The Random Forest and XGBoost models were evaluated using accuracy, precision, recall, F1-score, and AUC. After applying SMOTE, Random Forest achieved an accuracy of 77.66%, with a recall of 33.33% and an AUC of 0.67. XGBoost with SMOTE showed an accuracy of 82.31% and recall of 23.81%, with an AUC of 0.69. The application of SMOTE-ENN improved recall to 53.33% for Random Forest, while XGBoost reached a recall of 47.62%. Threshold tuning enhanced recall to 61.90%, but decreased precision to 22.57%. The Balanced Random Forest model, after hyperparameter tuning, achieved an accuracy of 79.31% and a recall of 31.43%. This work’s key contribution is the integration of XAI techniques to enhance model transparency, improving its applicability in clinical settings. Despite improvements, further optimization of precision and recall is needed for real-world deployment, and future research should focus on validating this framework across diverse datasets.
Bridging the Language of Cybersecurity: A Semantic Comparison Between NIST and ISO/IEC 27000 Terminologies Eka Julianti; Dewi Andriyanti; Aji Nurrohman; Rudolf Sinaga
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 2 (2026): June: Global Science: Journal of Information Technology and Computer Science
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i2.203

Abstract

Abstract: Background: Inconsistent terminology across cybersecurity frameworks undermines global governance and interoperability. The National Institute of Standards and Technology Cybersecurity Framework (NIST CSF 2.0) and ISO/IEC 27001:2022 share similar objectives but diverge semantically in defining risk, control, and resilience. This semantic gap causes difficulties in compliance mapping and automated policy translation. Research Objectives: This study aims to analyze the semantic similarity and divergence between NIST and ISO/IEC 27000 terminologies, identify conceptual structures influencing interoperability, and propose an AI-assisted foundation for harmonizing cybersecurity language globally. Methodology: A mixed-method semantic comparative design integrates Natural Language Processing (NLP) and ontology mapping. Using the nist_glossary.csv dataset and ISO vocabularies, terms were normalized and analyzed via cosine similarity using sentence-transformer embeddings. Ontological alignment was visualized through the Semantic Threat Graph (STG) and validated by certified experts using Cohen’s Kappa reliability tests. Results: From 672 term pairs, results show 40.9% high semantic equivalence, 38.8% partial overlap, and 20.3% semantic divergence. Strongest alignment appears in “Protect” and “Identify” domains, while divergences occur in governance and recovery-related terms. Ontology mapping revealed three conceptual clusters—Risk Governance, Technical Safeguards, and Organizational Readiness. Conclusions: Findings confirm a 79.7% total semantic alignment, indicating strong potential for harmonizing global cybersecurity standards. The study contributes an empirical model combining computational linguistics and AI-based ontology mapping to establish semantic interoperability, enabling unified cybersecurity governance and AI-driven compliance automation. Keywords: Semantic Interoperability; Ontology Mapping; Cybersecurity Frameworks; NIST; ISO/IEC 27000 Terminology
Predictive Modelling of Disease Patterns Using Time-Series Patient Data in Primary Healthcare Moh. Khoridatul Huda; Jerhi Wahyu Fernanda; Darmatasia
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 2 (2026): June: Global Science: Journal of Information Technology and Computer Science
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i2.204

Abstract

Understanding and predicting disease distribution patterns in primary healthcare settings require models capable of integrating both spatial and temporal dimensions. Traditional statistical approaches often fail to capture complex non-linear relationships across locations and time, leading to delayed detection of disease clusters. Objective: This study aims to develop a spatiotemporal machine learning framework to identify and forecast potential disease hotspots using electronic primary care records from 2024. Methods: The dataset comprised 5,343 patient visit records containing temporal, geographic (village-level), and clinical attributes. Data preprocessing included temporal aggregation, spatial encoding, and feature normalization. Three models—Gradient Boosting Machine (GBM), Temporal Random Forest (TRF), and Multi-EigenSpot—were trained and evaluated. Model performance was assessed using AUC, F1-score, and spatial accuracy metrics to ensure both predictive precision and spatial coherence. Results: Analysis revealed a clear seasonal pattern, with disease incidence peaking between April and August. Spatial mapping identified consistent hotspots in Sungai Asam and Beringin, accounting for over 70% of total cases. Among all tested models, Multi-EigenSpot achieved the best performance (AUC = 0.91; F1 = 0.86), effectively capturing multi-cluster spatial variability across months. Conclusions & Implications: The findings demonstrate that spatiotemporal learning models can significantly enhance disease surveillance and early warning capabilities in primary healthcare systems. Integrating spatial intelligence with explainable machine learning improves predictive accuracy, supports evidence-based policy, and enables targeted interventions for emerging disease hotspots in resource-limited settings.
Uncovering Latent Patient Profiles in Multidimensional Healthcare Data Using Unsupervised Clustering: A Comparative Analysis Waliya Rahmawanti; Ni Kadek Winda Patrianingsih; Usep Abdul Rosid
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 2 (2026): June: Global Science: Journal of Information Technology and Computer Science
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i2.205

Abstract

The increasing complexity of healthcare data presents significant challenges in identifying meaningful patient patterns, particularly in chronic disease management such as diabetes. Traditional analytical approaches often fail to capture hidden structures within multidimensional clinical data. This study aims to uncover latent patient profiles by applying unsupervised learning techniques to integrated demographic and physiological attributes, including age, body mass index, blood pressure, and gender. A comparative clustering framework was employed using K-Means, Hierarchical Clustering, and DBSCAN, with performance evaluated through Silhouette Score and Davies–Bouldin Index. The experimental results indicate that the optimal clustering configuration is achieved with K=3, where K-Means outperforms the other methods by producing more compact and well-separated clusters. Visual validation using Principal Component Analysis further confirms the structural coherence of the clusters. Specifically, Cluster 0 represents 60.4% of the dataset, with an average BMI of 24.31 and systolic blood pressure of 125.77 mmHg. Cluster 1, containing 29.3% of the patients, exhibits a higher average age of 61.16 years, with stable BMI and blood pressure values. Cluster 2, accounting for 10.2% of the patients, shows elevated physiological characteristics, including an average BMI of 26.49 and systolic pressure of 159.55 mmHg. The findings reveal three distinct patient groups: a majority group with moderate physiological conditions, an older-age group with relatively stable health indicators, and a smaller high-risk group characterized by elevated body mass index and significantly higher blood pressure levels. These results demonstrate the capability of unsupervised learning to identify clinically meaningful subpopulations without labeled data. In conclusion, this study provides a robust and scalable framework for patient segmentation using multidimensional healthcare data. The approach supports data-driven insights for healthcare analysis and risk stratification. Future work should explore the integration of additional clinical variables and advanced clustering techniques to enhance model interpretability and performance.
Data-Driven Analysis of Healthcare Service Efficiency Using Temporal and Process Mining Approaches Eka Julianti; Renny Afriany; Samsinar
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 3 (2026): September: Global Science: Journal of Information Technology and Computer Scien
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i3.227

Abstract

Background: Healthcare service efficiency has become a critical concern due to increasing patient demand and limited resources, where prolonged waiting time is a key indicator of inefficiency and reduced service quality. However, existing studies predominantly focus on prediction or aggregate analysis without integrating temporal characteristics and process-oriented perspectives to explicitly identify bottlenecks and patient heterogeneity. Objective:This study aims to evaluate healthcare service efficiency through a data-driven framework that integrates temporal analysis, bottleneck identification, and clustering techniques. Methods: The study utilizes 3,335 patient records and analyzes waiting time and service time using descriptive statistics, distribution analysis, rule-based bottleneck classification, and K-Means clustering with data normalization. Results: The results show a significant imbalance between waiting time (mean = 40.03 minutes; max = 1428; Standard Deviation (SD) = 103.01) and service time (mean = 4.45 minutes; max = 59; SD = 7.88), indicating high variability and extreme delays. Bottleneck analysis reveals that 79.7% of cases are dominated by waiting time, while only 20.3% are related to service processes. Clustering identifies three distinct patterns: efficient service (19.34 min waiting), queue bottleneck (505.25 min waiting), and complex service (29.70 min service time). Conclusions & Implications: This study contributes by providing an integrated analytical framework that uncovers both structural inefficiencies and heterogeneous patient patterns. In conclusion, inefficiency is predominantly driven by queue-related delays, highlighting the importance of optimizing patient flow management rather than solely improving clinical capacity.
Data-Driven School Phenotyping from Multidimensional Educational Indicators: PCA-Based Clustering and Stability Assessment Imeldawaty Gultom; Dedi Candro Parulian Sinaga; Marcelino Caetano Noronha
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 3 (2026): September: Global Science: Journal of Information Technology and Computer Scien
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i3.228

Abstract

Educational datasets increasingly contain multidimensional indicators describing school resources, participation, learning-related conditions, and institutional characteristics. When these indicators are analyzed simultaneously, direct clustering can be affected by dimensionality and correlated variables. This study develops a data-driven framework for identifying and assessing school phenotypes from multidimensional educational indicators using principal component analysis (PCA), clustering, and repeated-subsampling stability assessment. The analysis used the 2025 Rapor Publik Asesmen Nasional dataset for teachers in SD/MI/equivalent schools, comprising 283,948 observations and 92 variables. Twelve variables with complete missingness were removed, leaving 80 variables. A set of 33 core features was subsequently constructed, after which the administrative variable kd_kokab was excluded from the final analytical feature set, resulting in 32 features. Following imputation and scaling, PCA reduced the 32-dimensional feature space to seven components explaining 59.5413% of the total variance. K-means clustering applied to the seven-dimensional representation produced two clusters containing 114,399 (40.29%) and 169,549 (59.71%) observations, respectively. The final clustering yielded a Silhouette Score of 0.268125 and a Calinski–Harabasz index of 11,946.661. Stability assessment based on ten repeated samples of 30,000 observations produced a mean adjusted Rand index (ARI) of 0.989183 (SD = 0.006497), with ARI values ranging from 0.979404 to 0.997731. The findings establish a reproducible data-driven structure of two school groups and demonstrate high assignment stability under repeated subsampling. The framework provides an empirical basis for subsequent characterization of the indicators that differentiate the identified school profiles.
Measuring the Effectiveness of Federated Learning in Protecting User Privacy: Literature Study Ahmad Asyhadi; Ronald Naibaho
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 3 (2026): September: Global Science: Journal of Information Technology and Computer Scien
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i3.231

Abstract

Studi literatur ini mengidentifikasi efektivitas Federated Learning dalam menjaga privasi pengguna, mengatasi masalah-masalah yang berkembang seputar perlindungan data dalam sistem terdistribusi. Penelitian ini secara komprehensif meninjau berbagai metrik, model yang digunakan untuk mengevaluasi mekanisme perlindungan privasi dalam Federated Learning, menyoroti pentingnya metrik dalam menilai efektivitas model. Temuan utama meliputi Teknik Perlindungan Privasi, dimana studi ini mengidentifikasi beberapa metode dan teknik yang digunakan di Federated Learning untuk meningkatkan privasi pengguna, seperti Privasi Diferensial Lokal (LDP), Enkripsi Gradien, dan Enkripsi Homomorfik. Teknik-teknik ini terbukti memberikan perlindungan privasi yang kuat sambil mempertahankan utilitas data. Tantangan dalam Implementasi, yaitu analisis mengungkapkan tantangan signifikan yang dihadapi selama implementasi FL, termasuk kebutuhan metrik yang terdefinisi dengan baik untuk mengevaluasi perlindungan privasi secara efektif. Studi ini membahas metrik, model penting untuk mengukur efektivitas privasi, termasuk tingkat kebocoran privasi, pertukaran antara privasi dan utilitas model, dan ketahanan terhadap serangan seperti inversi model. Ini penting untuk memahami keseimbangan antara privasi dan kinerja dalam sistem Federated Learning. Hasil studi literatur menyimpulkan bahwa Federated Learning lebih unggul dibandingkan metode pembelajaran mesin tradisional dalam melindungi privasi pengguna dengan menjaga data tetap lokal dan hanya mengirimkan pembaruan model, sehingga mengurangi risiko pelanggaran data. Secara keseluruhan, studi literatur ini memberikan wawasan berharga tentang metodologi dan tantangan penerapan Pembelajaran Federasi untuk perlindungan privasi, serta menawarkan dasar untuk penelitian berikutnya dan aplikasi praktis di bidang keamanan data.
Artikel A Controlled Evaluation of URL Preprocessing Strategies for Smishing Detection Using TF-IDF and Linear SVM.: A Controlled Evaluation of URL Preprocessing Strategies for Smishing Detection Using TF-IDF and Linear SVM. Noni Rahmawati; Teuku Djauhari; Suroto
Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 3 (2026): September: Global Science: Journal of Information Technology and Computer Scien
Publisher : International Forum of Researchers and Lecturers

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70062/globalscience.v2i3.233

Abstract

URLs are common indicators in smishing messages, yet their treatment during text preprocessing can alter the information available to a classifier. Existing studies often use URLs as features or additional security evidence, while the effect of retaining, removing, or abstracting URLs within the same text-classification pipeline remains less clearly isolated. Objective: This study evaluates how three URL preprocessing strategies affect three-class SMS classification and smishing recognition. Methods: A public SMS phishing dataset containing 5,971 messages was audited for label inconsistencies and duplicates, resulting in 5,797 messages classified as ham, spam, or smishing. URLs were processed under three scenarios: retained in their original form, removed, or replaced with urltoken. All scenarios used identical word-level TF-IDF features, Linear SVM parameters, and five-fold group-aware cross-validation. Macro-F1 was the primary metric, complemented by class-specific smishing measures, URL-bearing subset analysis, prediction-transition analysis, and exact McNemar tests. Results: URL abstraction achieved the highest overall Macro-F1 of 0.9007, while URL retention produced the highest smishing F1 of 0.9082. Among 206 URL-bearing messages, URL retention achieved the highest Macro-F1 of 0.7422 and smishing F1 of 0.8466. Removing URLs reduced smishing recall from 0.8421 to 0.7684 and increased smishing-to-ham errors. Pairwise differences in classification correctness were not statistically significant. Conclusions: URL preprocessing has limited effect on aggregate performance but more clearly influences smishing recognition among messages containing URLs. Complete URL removal discards useful link-related information, whereas URL abstraction preserves link presence and raw retention preserves additional lexical cues. URL handling should therefore be treated as an explicit experimental design choice rather than a routine cleaning step.

Page 4 of 4 | Total Record : 39

 1   2   3   4 

Filter by Year

2025 2026


Reset
Filter By Issues
All Issue Vol. 2 No. 3 (2026): September: Global Science: Journal of Information Technology and Computer Scien Vol. 2 No. 2 (2026): June: Global Science: Journal of Information Technology and Computer Science Vol. 2 No. 1 (2026): March: Global Science: Journal of Information Technology and Computer Science Vol. 1 No. 4 (2025): December: Global Science: Journal of Information Technology and Computer Scienc Vol. 1 No. 3 (2025): September: Global Science: Journal of Information Technology and Computer Scien Vol. 1 No. 2 (2025): June: Global Science: Journal of Information Technology and Computer Science Vol. 1 No. 1 (2025): March : Global Science: Journal of Information Technology and Computer Science