Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal JOINCS (Journal of Informatics, Network, and Computer Science)
Jabbar Ahmed, Ghazwan
Unknown Affiliation
Computer Science & IT

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

Found 2 Documents
Search

 1 
Ensemble Deep Learning Strategy for Handling Imbalanced Credit Card Fraud Data: Strategi Pembelajaran Mendalam Ensemble untuk Menangani Data Penipuan Kartu Kredit yang Tidak Seimbang Hassan Mohammed, Zainab; Hatem Khorsheed, Farah; Jabbar Ahmed, Ghazwan
JOINCS (Journal of Informatics, Network, and Computer Science) Vol. 8 No. 2 (2025): November
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21070/joincs.v8i2.1670

Abstract

Credit card fraud remains a major challenge in the financial sector due to its dynamic nature and highly imbalanced transaction data. This study presents a robust deep ensemble learning approach that integrates spatial, sequential, and temporal learning capabilities. A series of preprocessing steps were applied, including feature normalization, class-label separation, and class rebalancing using SMOTE. The model architecture combines convolutional, recurrent, and long short-term memory layers to capture diverse fraud patterns. These components are merged and passed through dense and dropout layers for optimal binary classification. The datasets used are generated from real-world credit card transactions, ensuring practical relevance. On the test set, the proposed model achieved 99.7% accuracy, 99.6% precision, 99.9% recall, and 99.8% F1-score. The training and validation loss curves showed smooth convergence without any overfitting, confirming model stability. To ensure reliability, 3-fold stratified cross-validation was performed on the balanced dataset. The average metrics across folds included 99.76% accuracy, 99.70% precision, 99.85% recall, and 99.77% F1-score. These results underscore the generalization capability and consistent prediction performance of the model. Comparative analysis showed that the group model outperformed individual CNN, RNN, and LSTM architectures. The hybrid strategy benefits from the spatial extraction of CNN, sequence modeling of RNN, and memory retention of LSTM. By integrating these strengths, the model effectively detects subtle and complex fraud patterns. This approach provides a scalable and reliable solution for real-time fraud detection in imbalanced credit card datasets.
Designing an Assistive Tool for Visually Impaired People Based on Object Detection Technique: Merancang Alat Bantu Bagi Penyandang Disabilitas Visual Berbasis Teknik Deteksi Objek Jabbar Ahmed, Ghazwan; Hatem Khorsheed, Farah; Kadhim Zaidan, Fadhil
JOINCS (Journal of Informatics, Network, and Computer Science) Vol. 8 No. 2 (2025): November
Publisher : Universitas Muhammadiyah Sidoarjo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21070/joincs.v8i2.1672

Abstract

Visually impaired individuals often face significant challenges in navigating their environments due to limited access to visual information. To address this issue, we propose an assistive tool designed to operate on a PC. The focus of this research is on developing an efficient, lightweight object detection system to ensure real-time performance while maintaining compatibility with low-resource setups. The proposed system enhances the autonomy and accessibility of visually impaired individuals by providing audio descriptions of their surroundings through the processing of live-streaming video. The core of the system is an object detection module based on the state-of-the-art YOLO7 model, designed to identify multiple objects in real-time within the user's environment. The system processes video frames captured by a camera, identifies objects, and delivers the results as audio descriptions using the pyttsx3 text-to-speech library, ensuring offline functionality and robust performance. The system demonstrates satisfactory results, achieving inference speeds ranging from 0.12 to 1.14 seconds for object detection, as evaluated through quantitative metrics and subjective assessments. In conclusion, the proposed tool effectively aids visually impaired individuals by providing accurate and timely audio descriptions, thereby promoting greater independence and accessibility.
Co-Authors Hassan Mohammed, Zainab Hatem Khorsheed, Farah Kadhim Zaidan, Fadhil