Garuda - Garba Rujukan Digital

Stacked LSTM with Multi Head Attention Based Model for Intrusion Detection S Phani Praveen; Padmavathi Panguluri; Uddagiri Sirisha; Deshinta Arrova Dewi; Tri Basuki Kurniawan; Lusiana Efrizoni
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.764

The rapid advancement of digital technologies, including the Internet of Things (IoT), cloud computing, and mobile communications, has intensified reliance on interconnected networks, thereby increasing exposure to diverse cyber threats. Intrusion Detection Systems (IDS) are essential for identifying and mitigating these threats; however, traditional signature-based and rule-based methods fail to detect unknown or complex attacks and often generate high false positive rates. Recent studies have explored machine learning (ML) and deep learning (DL) approaches for IDS development, yet many suffer from poor generalization, limited scalability, and an inability to capture both spatial and temporal dependencies in network traffic. To overcome these challenges, this study proposes a hybrid deep learning framework integrating Convolutional Neural Networks (CNN), Stacked Long Short-Term Memory (LSTM) networks, and a Multi-Head Self-Attention (MHSA) mechanism. CNN layers extract spatial features, stacked LSTM layers capture long-term temporal dependencies, and MHSA enhances focus on the most relevant time steps, improving accuracy and reducing false alarms. The proposed model was trained and evaluated on the UNSW-NB15 dataset, which represents modern attack vectors and realistic network behavior. Experimental results show that the model achieves state-of-the-art performance, attaining 99.99% accuracy and outperforming existing ML and DL-based intrusion detection systems in both precision and generalization capability.

USING AI TECHNOLOGY TO SUPPORT LEARNING AND BOOST STUDENT INDEPENDENCE AT SMA MUHAMMADIYAH PK SURAKARTA Muhammad Aliq Aulia; Kresno Ario Tri Wibowo; Irfan Nugraha; Deshinta Arrova Dewi
Journal of Community Service Vol 8 No 1 (2026): JCS, June 2026
Publisher : Ikatan Dosen Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56670/jcs.v8i1.399

Background: The rapid development of Artificial Intelligence (AI) technology has significantly influenced the education sector, particularly at the senior high school level. However, preliminary observations at SMK Muhammadiyah PK Surakarta indicated that students’ understanding of AI was still limited prior to the implementation of this community service activity. Initial assessments showed that only about 20% of students had basic knowledge of AI concepts, including its functions, limitations, and ethical use in learning contexts. This gap between technology exposure and conceptual understanding highlights the importance of structured AI socialization to support learning and foster student independence. Objective: This community service activity aims to support learning processes and boost student independence by strengthening AI literacy through systematic socialization and educational activities. Method: The program was implemented through AI socialization sessions, interactive presentations, demonstrations of AI-based learning tools, and guided discussions. The activity was carried out through an international collaboration with INTI International University, with academic partnership support from Prof. Dr. Deshinta Arrova Dewi. The evaluation employed a one-group pre-test and post-test design to measure changes in students’ understanding of AI. Result: The post-test results indicated a substantial improvement in students’ understanding of Artificial Intelligence. The level of comprehension increased from approximately 20% before the socialization to 89% after the activity, reflecting an overall improvement of 69%. This improvement demonstrates that AI socialization effectively enhances students’ digital literacy, learning engagement, and capacity for independent learning. Conclusion: These findings indicate that AI socialization is essential and effective for senior high school students. The community service activity focusing on Using AI Technology to Support Learning and Boost Student Independence can be well accepted by the target participants and has strong potential to be implemented sustainably as an innovative learning support model in secondary education.

Incorporate Transformer-Based Models for Anomaly Detection Deshinta Arrova Dewi; Harprith Kaur Rajinder Singh; Jeyarani Periasamy; Tri Basuki Kurniawan; Henderi Henderi; M. Said Hasibuan; Yogeswaran Nathan
Journal of Applied Data Sciences Vol 6, No 3: September 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i3.762

This paper explores the effectiveness of Transformer-based models, specifically the Time-Series Transformer (TST) and Temporal Fusion Transformer (TFT), for anomaly detection in streaming data. We review related work on anomaly detection models, highlighting traditional methods' limitations in speed, accuracy, and scalability. While LSTM Autoencoders are known for their ability to capture temporal patterns, they suffer from high memory consumption and slower inference times. Though efficient in terms of memory usage, the Matrix Profile provides lower performance in detecting anomalies. To address these challenges, we propose using Transformer-based models, which leverage the self-attention mechanism to capture long-range dependencies in data, process sequences in parallel, and achieve superior performance in both accuracy and efficiency. Our experiments show that TFT outperforms the other models with an F1-score of 0.92 and a Precision-Recall AUC of 0.71, demonstrating significant improvements in anomaly detection. The TST model also shows competitive performance with an F1-score of 0.88 and Precision-Recall AUC of 0.68, offering a more efficient alternative to LSTMs. The results underscore that Transformer models, particularly TST and TFT, provide a robust solution for anomaly detection in real-time applications, offering improved performance, faster inference times, and lower memory usage than traditional models. In conclusion, Transformer-based models stand out as the most effective and scalable solution for large-scale, real-time anomaly detection in streaming time-series data, paving the way for their broader application across various industries. Future work will further focus on optimizing these models and exploring hybrid approaches to enhance detection capabilities and real-time performance.

A Proposed Model for Detecting Learning Styles Based on the Felder-Silverman Model Using KNN and LR with Electroencephalography (EEG) Muhammad Said Hasibuan; R Rizal Isnanto; Deshinta Arrova Dewi; Tri Basuki Kurniawan; Ming-Lang Yeh; Adi Wijaya
Journal of Applied Data Sciences Vol 6, No 2: MAY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i2.659

The identification of learning styles plays a crucial role in enhancing personalized education and optimizing learning outcomes. This research proposes a model for detecting learning styles based on the Felder-Silverman model using two machine learning algorithms: K-Nearest Neighbors (KNN) and Linear Regression (LR). Electroencephalography (EEG) data, known for its ability to capture cognitive and neural activity, serves as the primary dataset for this study. The proposed model was tested on a dataset comprising EEG signals collected during various learning tasks. Feature extraction and preprocessing techniques were employed to ensure high-quality input for the learning algorithms. The experimental results revealed that the LR-based model achieved an accuracy of 96.4%, significantly outperforming the KNN-based model, which obtained an accuracy of 89.9%. These findings highlight the potential of EEG-based models for accurately identifying learning styles, offering valuable insights for educators and researchers aiming to implement adaptive learning systems. This study demonstrates the feasibility and effectiveness of combining EEG data with machine learning techniques for learning style detection, paving the way for more personalized and efficient educational approaches. Future research will explore the integration of additional physiological data and advanced machine learning methods to further improve model accuracy and applicability.

Detecting Gender-Based Violence Discourse Using Deep Learning: A CNN-LSTM Hybrid Model Approach Tri Basuki Kurniawan; Deshinta Arrova Dewi; Henderi Henderi; M. Said Hasibuan; Mohd Zaki Zakaria; Abdul Azim Bin Ismail
Journal of Applied Data Sciences Vol 6, No 3: September 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i3.761

Gender-Based Violence (GBV) is a critical social issue impacting millions worldwide. Social media discussions offer valuable insights into public awareness, sentiment, and advocacy, yet manually analyzing such vast textual data is highly challenging. Traditional text classification methods often struggle with contextual understanding and multi-class categorization, making it difficult to accurately identify discussions on Sexual Violence, Physical Violence, and other topics. To address this, the present study proposes a hybrid deep learning approach combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. CNN is utilized for extracting key linguistic features, while LSTM enhances the classification process by maintaining sequential dependencies. This hybrid CNN+LSTM model is evaluated against standalone CNN and LSTM models to assess its performance in classifying GBV-related tweets. The dataset was sourced from Kaggle, containing real-world Twitter discussions on GBV. Experimental results demonstrate that the hybrid model surpasses both CNN and LSTM models, achieving an accuracy of 89.6%, precision of 88.4%, recall of 89.1%, and F1-score of 88.7%. Confusion matrix and ROC curve analyses further confirm the hybrid model’s superior performance, correctly identifying Sexual Violence (82%), Physical Violence (15%), and Other (3%) cases with reduced misclassification rates. These results suggest that combining CNN’s feature extraction with LSTM’s contextual learning provides a more balanced and effective classification model for GBV-related text. This work supports the development of AI-based tools for social media monitoring, policy-making, and advocacy, helping stakeholders better understand and respond to GBV discussions. Future research could explore transformer-based models like BERT and real-time classification applications to further improve performance.

Navigating Heart Stroke Terrain: A Cutting-Edge Feed-Forward Neural Network Expedition S Phani Praveen; Jeevana Sujitha Mantena; Uddagiri Sirisha; Deshinta Arrova Dewi; Tri Basuki Kurniawan; Choo Wou Onn; Yorman Yorman
Journal of Applied Data Sciences Vol 6, No 3: September 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i3.763

Heart stroke remains one of the leading causes of death worldwide, necessitating early and accurate prediction systems to enable timely medical intervention. While a variety of machine learning approaches have been employed to address this issue, including Logistic Regression, Decision Trees, Random Forests, Support Vector Machines, and K-Nearest Neighbors, these models often suffer from limitations such as overfitting, insufficient generalization, poor performance on imbalanced datasets, and inability to capture complex nonlinear patterns in clinical data. Additionally, many existing works do not comprehensively integrate both clinical and demographic features or lack rigorous evaluation metrics beyond accuracy alone. This study proposes a novel Feed-Forward Neural Network (FFNN) model for heart stroke prediction, designed to overcome the shortcomings of conventional models. Unlike shallow classifiers, the FFNN architecture employed here leverages multiple hidden layers and nonlinear activation functions to learn intricate relationships within the dataset. The dataset used comprises various attributes such as age, hypertension, heart disease, BMI, and smoking status, which were preprocessed through normalization, one-hot encoding, and imputation techniques to ensure data quality and model performance. Experiments were conducted using a stratified train-test split, and the model was trained using the Adam optimizer with carefully tuned hyperparameters. Comparative evaluations against baseline models (Logistic Regression, Random Forest, and SVM) were carried out using precision, recall, F1-score, and ROC-AUC as performance metrics. The proposed FFNN achieved the highest accuracy of 96.47%, along with substantial improvements in recall and F1-score, highlighting its superior capability in identifying potential stroke cases even in imbalanced datasets. This work bridges a significant gap in heart stroke prediction by demonstrating the effectiveness of deep learning models—specifically FFNNs—in extracting complex patterns from diverse patient data. It also sets the stage for further exploration of deep learning-based clinical decision support systems.

Deep Learning Based Face Mask Detection System Using MobileNetV2 for Enhanced Health Protocol Compliance Fadly Fadly; Tri Basuki Kurniawan; Deshinta Arrova Dewi; Mohd Zaki Zakaria; Putri Aisha Athira binti Hisham
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.476

Personal protective equipment (PPE) is crucial in mitigating the spread of infections within the pharmacy industry, manufacturing sectors, and healthcare facilities. Airborne particles and contaminants can be released during the handling of pharmaceuticals, the operation of machinery, or patient care activities. These particles can be transmitted through close contact with an infected individual or by touching contaminated surfaces and then touching one's face (mouth, nose, or eyes). PPE, including face masks, plays a vital role in minimizing the risk of transmission of infectious diseases. Although mandates for wearing face masks might relax as situations improve and vaccination rates increase, staying prepared for potential future outbreaks and the resurgence of infectious diseases remains important. Therefore, an automated system for face mask detection is important for future use. This research proposes real-time face mask detection by identifying who is (i) not wearing a mask and (ii) wearing a mask. This research presents a deep-learning approach using a pre-trained model, MobileNet-V2. The model is trained on a 10,000 dataset of images of individuals with and without masks. The result shows that the pre-trained MobileNet-V2 model obtained a high accuracy of 98.69% on the testing dataset.

Fake vs Real Image Detection Using Deep Learning Algorithm Fatoni Fatoni; Tri Basuki Kurniawan; Deshinta Arrova Dewi; Mohd Zaki Zakaria; Abdul Muniif Mohd Muhayeddin
Journal of Applied Data Sciences Vol 6, No 1: JANUARY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i1.490

The purpose of this research project is to address the growing issues presented by modified visual information by developing a deep learning model for identifying between real and fake images. To enhance accuracy, this project evaluates the effectiveness of deep learning algorithms such as Residual Neural Network (ResNet), Visual Geometry Group 16 (VGG16), and Convolutional Neural Network (CNN) together with Error Level Analysis (ELA) as preprocessing the dataset. The CASIA dataset contains 7,492 real images and 5,124 fake images. The images included are from a wide range of random subjects, including buildings, fruits, animals, and more, providing a comprehensive dataset for model training and validation. This research examined models' effectiveness through experiments, measuring their training and validation accuracies. It comes out with the best accuracy of each model, which is for Convolutional Neural Network (CNN), 94% for training accuracy, and validation accuracy of 92%. For VGG16, with both training and validation accuracy reaching 94%. Lastly, Residual Neural Network (ResNet) demonstrated optimal performance with 95% training accuracy and 93% validation accuracy. This project also constructs a system prototype for practical applications, offering an interface for real-world testing. When integrating into the system prototype, only Residual Neural Network (ResNet) shows consistency and effectiveness when predicting both fake and real images, and this led to the decision to choose ResNet for integration into the system. Furthermore, the project identified several areas for improvement. Firstly, expanding the model comparison for discovering more successful algorithms. Next, improving the dataset preprocessing phase by incorporating filtering or denoising techniques. Lastly, refining the system prototype for greater appeal and user-friendliness has the potential to attract a larger audience.

Stacked LSTM with Multi Head Attention Based Model for Intrusion Detection S Phani Praveen; Padmavathi Panguluri; Uddagiri Sirisha; Deshinta Arrova Dewi; Tri Basuki Kurniawan; Lusiana Efrizoni
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.764

The rapid advancement of digital technologies, including the Internet of Things (IoT), cloud computing, and mobile communications, has intensified reliance on interconnected networks, thereby increasing exposure to diverse cyber threats. Intrusion Detection Systems (IDS) are essential for identifying and mitigating these threats; however, traditional signature-based and rule-based methods fail to detect unknown or complex attacks and often generate high false positive rates. Recent studies have explored machine learning (ML) and deep learning (DL) approaches for IDS development, yet many suffer from poor generalization, limited scalability, and an inability to capture both spatial and temporal dependencies in network traffic. To overcome these challenges, this study proposes a hybrid deep learning framework integrating Convolutional Neural Networks (CNN), Stacked Long Short-Term Memory (LSTM) networks, and a Multi-Head Self-Attention (MHSA) mechanism. CNN layers extract spatial features, stacked LSTM layers capture long-term temporal dependencies, and MHSA enhances focus on the most relevant time steps, improving accuracy and reducing false alarms. The proposed model was trained and evaluated on the UNSW-NB15 dataset, which represents modern attack vectors and realistic network behavior. Experimental results show that the model achieves state-of-the-art performance, attaining 99.99% accuracy and outperforming existing ML and DL-based intrusion detection systems in both precision and generalization capability.

Progressive Massive Fibrosis Detection Using Generative Adversarial Networks and Long Short-Term Memory Suhendro Y. Irianto; Sri Karnila; M.S. Hasibuan; Deshinta Arrova Dewi; Tri Basuki Kurniawan; Hendra Kurniawan
Journal of Applied Data Sciences Vol 6, No 4: December 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i4.707

Contribution: Progressive Massive Fibrosis (PMF) is a severe form of pneumoconiosis, affecting individuals exposed to mineral dust, such as coal miners and workers in the artificial stone industry. This condition causes significant pulmonary impairment and increased mortality. Early and accurate detection is vital for effective management, yet traditional diagnostic methods face challenges in differentiating PMF from other pulmonary diseases due to variability in clinical presentations and limitations in imaging techniques. Idea: The study introduces a novel diagnostic framework that integrates Generative Adversarial Networks (GAN) and Long Short-Term Memory (LSTM) networks to enhance the detection and monitoring of PMF. The GAN generates high-fidelity synthetic imaging data to address the issue of limited datasets, while the LSTM network captures temporal patterns in patient data, enabling real-time monitoring of disease progression. Objective: The primary objective of this research is to develop an AI-driven model that improves the accuracy and efficiency of PMF detection and monitoring, facilitating early diagnosis and better treatment planning. Findings: The integrated GAN-LSTM model significantly outperformed traditional diagnostic methods. It proved high accuracy, a Dice coefficient of 0.85, and an Area Under the Curve (AUC) of 0.92, showing precise differentiation of PMF from other pulmonary conditions, such as lung cancer and tuberculosis. Results: The GAN-LSTM framework achieved an accuracy of 91.3%, suggesting that the fusion of GAN and LSTM technologies can effectively address the challenges of limited datasets and heterogeneous disease progression. The model showed promise in enhancing the non-invasive detection and ongoing monitoring of PMF. Novelty: This research stands for a significant advancement in PMF diagnostics by combining GAN and LSTM technologies in a single framework. This approach improves diagnostic accuracy and eases continuous disease monitoring, offering a non-invasive and highly precise solution for PMF detection.

Title

Found 11 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 11 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 11 Documents
Search