Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Jurnal Kridatama Sains dan Teknologi International Journal of Engineering, Technology and Natural Sciences (IJETS)
Yadhurani Dewi Amritha
Universitas Pendidikan Nasional, Denpasar, Indonesia
Agriculture, Biological Sciences & Forestry Civil Engineering, Building, Construction & Architecture Computer Science & IT Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Industrial & Manufacturing Engineering Social Sciences

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

Found 2 Documents
Search

 1 
Model Machine Learning yang Dioptimalkan untuk Prediksi Penyakit Jantung Menggunakan R Shiny Yadhurani Dewi Amritha; Ni Luh Putu Ika Candrawengi; Md Wira Putra Dananjaya; Made Ari Riska Dayanti
Jurnal Kridatama Sains dan Teknologi Vol 8 No 01 (2026): Jurnal Kridatama Sains dan Teknologi
Publisher : Universitas Ma'arif Nahdlatul Ulama Kebumen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53863/kst.v8i01.1994

Abstract

Heart disease continues to be a major contributor to global mortality, highlighting the critical importance of early detection in enhancing patient outcomes. The increasing availability of structured clinical datasets has enabled the application of intelligent systems for risk prediction and diagnostic support. In this paper, the effectiveness of three supervised learning algo- rithms—Random Forest (RF), Support Vector Machine (SVM), and Decision Tree (DT)—is evaluated for the task of heart disease prediction. This investigation is based on the Heart Failure Prediction dataset sourced from the Kaggle platform. The training process for each model involved a 10-fold cross- validation, with its hyperparameters later being tuned using grid search optimization. Model efficacy was measured against standard classification benchmarks, including accuracy, sensitivity, specificity, and the area under the ROC curve (AUC). The Random Forest model emerged as the most effective, demon- strating superior performance with an AUC of 0.9517, sensitivity of 81.18%, and specificity of 90.44%. To facilitate clinical use, this model was subsequently integrated into a user-friendly web tool built with the R Shiny framework. The interface allows users to input patient-level clinical data and obtain real-time predictions, along with visualizations of feature importance and risk probability. This implementation bridges the gap between algorithm development and practical application, offering a user- friendly decision support tool for early heart disease screening. The findings affirm that machine learning models, when properly tuned and validated, can serve as effective and interpretable tools in clinical decision-making. This work contributes to the advancement of e-health and the integration of AI-driven models into medical workflows
Human Intruder Detection System (IDS) for Restricted Security Area: A Systematic Literature Review Yadhurani Dewi Amritha; I Made Yogaswara Dipta
Jurnal Internasional Teknik, Teknologi dan Ilmu Pengetahuan Alam Vol 7 No 2 (2025): International Journal of Engineering, Technology and Natural Sciences
Publisher : Universitas Teknologi Yogyakarta, Yogyakarta, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46923/ijets.v7i2.457

Abstract

Ensuring security in sensitive areas such as airports, military bases, and nuclear facilities is critical to prevent unauthorized access. Traditional reliance on security personnel is often inefficient and insufficient for continuous monitoring. Intruder Detection Systems (IDS), which utilize devices or sensors to detect unauthorized entry, have emerged as essential tools for safeguarding high-security environments. However, there is a lack of comprehensive understanding that systematically synthesizes existing research on human intruder detection. This study aims to conduct a systematic literature review (SLR) on human IDS to provide a structured overview of current methodologies, technologies, and challenges in the field. Using established SLR protocols, relevant studies were collected, analyzed, and categorized to identify prevailing trends and gaps. The results highlight various object detection techniques and their effectiveness in real-world security applications. Despite the advances, challenges such as limited environmental adaptability and real-time accuracy remain. The findings of this review offer valuable insights for professionals and future researchers, guiding the development of more robust and efficient human intruder detection solutions.
Co-Authors I Made Yogaswara Dipta Made Ari Riska Dayanti Md Wira Putra Dananjaya Ni Luh Putu Ika Candrawengi