Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal INOVTEK Polbeng - Seri Informatika
Denar Regata Akbi
Muhammadiyah University of Malang
Computer Science & IT

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

Found 2 Documents
Search

 1 
Comparative Analysis of Performance and Interpretability of XGBoost and TabNet Models in IDS Using XAI Ahmad Jauharul Ilmi; Denar Regata Akbi
Journal of Innovation and Technology Polbeng Series on Informatics (INOVTEK Polbeng - Seri Informatika) Vol. 11 No. 3 (2026): August (Inpress)
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/js504z34

Abstract

Cyberattacks are becoming a more apparent danger to modern network traffic. Robust and transparent Intrusion Detection Systems (IDS) are increasingly needed to counter this massive wave. The emergence of machine learning (ML) and deep learning (DL) offers a huge advantage in detecting this wave with high precision; nevertheless, they also have a downside due to their secretive nature regarding decision-making algorithms, often referred to as the “black-box” problem. Digital forensics is affected by this nature as well. As a result, XGBoost (ML) and TabNet (DL) were selected and compared based on their performance and interpretability using Explainable AI (XAI) to understand the reasoning behind their decisions with the CIC-IDS-2017 dataset in this study. Both models naturally achieved extremely high classification capabilities, with XGBoost slightly outperforming TabNet in overall detection reliability and in minimizing false negatives. The XAI evaluation additionally uncovered equally valid decision-making logic: TabNet prioritizes connection states and temporal variances, while XGBoost primarily relies on rigid volumetric payload statistics. LIME analysis also affirms that both models maintain high consistency with SHAP global explanations. In summary, XGBoost is recommended as the primary real-time detector due to its superior precision and transparent logic, while TabNet serves as a secondary validator.
Shap and Lime Analysis on CNN-GRU Deep Learning Models for IoT Network Intrusion Detection Satria Purfie Purnama Putra; Denar Regata Akbi
Journal of Innovation and Technology Polbeng Series on Informatics (INOVTEK Polbeng - Seri Informatika) Vol. 11 No. 3 (2026): August (Inpress)
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/5r93tw26

Abstract

As digital devices evolve quickly and cyberattacks grow more diverse, the Internet of Things (IoT) ecosystem faces a much higher security risk than before. An AI-based security system is commonly used for post-attack mitigation. However, this approach has several issues, such as a high computational load, data imbalance, and a black box that does not explain the attack patterns. This study addresses a research gap by implementing a hybrid CNN-GRU architecture. With XAI, this architecture remains lightweight and robust while mapping out malicious attack patterns quite clearly. The technique breaks down these security incidents by providing both global and local explanations. This study uses two IoT datasets, BoT-IoT with 72 million records and IoTID20 with 625,783 records. The SMOTE post-split technique was performed on 80% of the total data to address data imbalance and avoid data leakage. To validate the results, stratified holdout is used to evaluate the training results. This research has very satisfying results with 99% accuracy and a 99% F1-score so that it can minimize errors in both datasets. The contributions made by this research are (1) adapting a stable model for anomaly classification, (2) handling data imbalance and avoiding data leakage, and (3) integrating SHAP and LIME to overcome black boxes.
Co-Authors Ahmad Jauharul Ilmi Satria Purfie Purnama Putra