Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Engineering, Mathematics and Computer Science Journal (EMACS)
Jabar, Bakti Amirul
Unknown Affiliation
Civil Engineering, Building, Construction & Architecture Computer Science & IT Engineering Industrial & Manufacturing Engineering Mathematics

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

Found 2 Documents
Search
Journal : Engineering, Mathematics and Computer Science Journal (EMACS)

 1 
Indoor Positioning System using Gaussian Mixture Model on BLE Fingerprint Lie, Maximilianus Maria Kolbe; Jabar, Bakti Amirul
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 7 No. 1 (2025): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v7i1.12581

Abstract

After the release of Bluetooth Low Energy (BLE), people have been trying to use Bluetooth as an alternative source to solve indoor positioning. Unfortunately, due to the nature of Bluetooth about proximity, the propagated signal is very fluctuating. This decreases the accuracy considerably and has become one of the main problems in using Bluetooth. To combat the signal fluctuations, we propose a fingerprinting-based concept of using received signal strength (RSS) frequency distribution values as the data in the radio map, which is termed Frequency Distribution Radio Map (FDRM). We also propose a probabilistic fingerprinting-based algorithm utilizing FDRM using Gaussian Mixture Model (GMM) as the probability distribution function (PDF). In the offline phase, we compare 2 methods: k-Means only, and k-Means with Expectation-Maximization (EM); to learn the FDRM. This resulting a probability distribution function (PDF) of the RSS in each reference points for each BLEs. In the online phase, k-Nearest Neighbour (KNN) and weighted average are used to estimate the receiver’s location. The proposed method is validated over 3 different datasets taken from a 4 m x 6 m classroom equipped with chairs and tables. The experiment shows that the proposed fingerprint and model are better in capturing the environment, including the signal fluctuation. By using only k-Means in obtaining the GMM, it achieved mean error of 98.18 cm and standard deviation of 56.11 cm. By adding EM, there will be a trade-off between mean error with better standard deviation and lower computing time. It achieved standard deviation of 47.99 cm and mean error of 112.24 cm.
A Comparative Study of Machine Learning and Stacking Ensemble Models for Diabetes Prediction Jabar, Bakti Amirul; Januario, Albertus; Sanjaya, Davin Miguel; Tanuwijaya, James
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 8 No. 1 (2026): EMACS (In Press)
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v8i1.15332

Abstract

Diabetes is a chronic metabolic disease and increasingly widespread disease around the world and early diagnosis is crucial. Methodology In this study, the performance of three machine learning models (Logistic- Regression, K-Nearest Neighbour (KNN) and Naive Bayes) is reviewed under the task of diabetes classification using Pima Indians Diabetes Dataset. To tackle the class imbalance, we applied imputation, SMOTE for the data pre-processing, and Min-Max Scaling to enhance the prediction performance. Further, we have applied the ensemble learning and stacking, where all the three models have been used as meta classifiers. The results indicate that KNN had the best individual model performance (accuracy 77.27%, AUC 0.8444%) but the stacking ensemble with meta-model being Logistic Regression is superior to any model (accuracy 80.52%, AUC 0.8604%). This suggests that ensemble learning can also improve the accuracy of diabetes diagnosis. These findings demonstrate that combining multiple classification approaches may provide more stable predictions across different patient conditions and clinical attributes In addition the preprocessing stages contributed to reducing noise and improving data consistency before model training The study also highlights the potential use of ensemble-based systems in supporting healthcare professionals during preliminary diabetes screening particularly in environments with limited medical resources and increasing numbers of diabetes cases requiring rapid assessment.
Co-Authors Januario, Albertus Lie, Maximilianus Maria Kolbe Sanjaya, Davin Miguel Tanuwijaya, James