Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Jurnal Teknologi Informatika dan Komputer Jurnal Visi Pengabdian Kepada Masyarakat
Dharmayanti
Unknown Affiliation
Arts Humanities Computer Science & IT Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Mathematics Mechanical Engineering Social Sciences

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
The Development of Medical Skills Through Human Resources Management in Griya Sehat Shareeah, Depok City Dona Suzana; Ditiya Himawati; Quroyzhin Kartika Rini; Dharmayanti
Jurnal Visi Pengabdian Kepada Masyarakat Vol. 6 No. 1 (2025): Jurnal Visi Pengabdian Kepada Masyarakat : Edisi Februari 2025
Publisher : LPPM Universitas HKBP Nommensen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51622/pengabdian.v6i1.2542

Abstract

In the era of globalization, traditional therapy clinics are required to ensure high personnel performance in providing services to the community, especially patients as the object of service. The low level of knowledge of health workers regarding health credential management affects the performance and quality of services in clinics. The aim of the health workers credential training activity is to improve the quality of health services at Griya Sehat Shareeah, Depok City. The community service activity was attended by 13 health workers from Griya Sehat Shareeah. Training is carried out with several practice sessions and an ongoing training program that focuses on technical and moral support efforts. The results of training show that there are differences in the level of knowledge after being given training to health workers. Thus, credential management training using lectures and practice methods can increase health workers knowledge about credential management and strengthen knowledge of the competence and authority of health workers at Griya Sehat Shareeah in providing health services so as to improve service quality. In the end, Griya Sehat Shareeah became an exemplary model that was able to combine traditional medicine with modern medicine in society.
Zero-Day Attack Detection Using Autoencoder and XGBoost Rohman, Mujibbur; Dharmayanti
Jurnal Teknologi Informatika dan Komputer Vol. 12 No. 1 (2026): Jurnal Teknologi Informatika dan Komputer
Publisher : Universitas Mohammad Husni Thamrin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37012/jtik.v12i1.3248

Abstract

Advances in information and communication technology have significantly impacted progress in various sectors, but they have also given rise to increasingly complex network security threats. Cyberattacks such as Distributed Denial of Service (DDoS), ransomware, and software vulnerability exploits continue to increase year after year. Signature-based Intrusion Detection Systems are often ineffective in identifying novel cyberattacks since they rely solely on previously known attack patterns. To address this limitation, this study proposes a hybrid approach that integrates Autoencoders, including Dense and Memory-Augmented variants, with Extreme Gradient Boosting (XGBoost) to enhance zero-day attack detection using the UNSW-NB15 dataset. The research methodology encompasses data exploration, preprocessing with a split-before-transform strategy to prevent information leakage, Autoencoder training to model normal network behavior, reconstruction error computation for anomaly detection under both fixed and adaptive thresholding, and the utilization of these errors as input features for XGBoost classification. Experimental results demonstrate that adaptive thresholding improves F1 performance compared to fixed thresholds, while the hybrid Autoencoder–XGBoost integration achieves a significant performance boost. The proposed model consistently obtained F1 scores above 0.80 and PR-AUC values exceeding 0.81 with a balanced trade-off between precision and recall. These findings confirm that the hybrid approach is more effective, consistent, and adaptive in detecting intrusions, particularly zero-day attacks, and highlight its potential as a robust framework for advancing network security in dynamic threat environments.
Co-Authors Ditiya Himawati Dona Suzana Quroyzhin Kartika Rini Rohman, Mujibbur