Publication Search

Explainable artificial intelligence (XAI) has become a critical requirement in cybersecurity due to the high-stakes nature of security decision-making and the limitations of black-box learning models. This study investigates the construction of an explainable cybersecurity knowledge representation by leveraging standardized terminology from the NIST cybersecurity glossary. The primary problem addressed is the lack of transparent and semantically grounded reasoning mechanisms in existing AI-driven cybersecurity systems, which limits trust, accountability, and analyst adoption. To address this challenge, we propose a NIST-based semantic knowledge graph that embeds explainability directly into its ontology structure and reasoning process. The proposed framework systematically extracts definitional entities and relations from NIST glossary entries to construct a domain ontology and a multi-relational knowledge graph. A rule-based semantic relation extraction method is employed to ensure faithful, interpretable, and reproducible reasoning paths. The resulting knowledge graph contains over 3,000 cybersecurity concepts and approximately 27,000 semantic relations, covering hierarchical, associative, dependency, and mitigation semantics. Experimental evaluation demonstrates that the proposed approach achieves a high level of explainability, with 92.4% of reasoning outcomes being fully traceable and only 1.4% classified as non-traceable. Most explainable reasoning paths are limited to two or three hops, indicating an effective balance between inferential depth and human interpretability. Structural analysis further confirms the presence of meaningful hub concepts that support multi-hop semantic inference. These results confirm that ontology-driven, standard-based knowledge graphs provide a robust foundation for explainable cybersecurity intelligence. The study concludes that explainability-by-design, grounded in authoritative standards, offers a viable and trustworthy alternative to opaque AI models for cybersecurity applications.

High-dimensional clinical data exhibit complex and non-linear relationships among patient attributes, where outcomes are often influenced by feature interactions rather than isolated variables. However, many existing machine learning models prioritize predictive performance while providing limited interpretability and insufficient insight into interaction structures. This study aims to address this limitation by developing an interpretable and robust framework for feature interaction mining in clinical data. We propose a hybrid tree–neural modeling framework that explicitly captures and ranks feature interactions while maintaining stable predictive performance. Tree-based ensemble models are employed to identify non-linear interaction patterns, while neural representations enhance learning flexibility and generalization. The framework integrates interaction importance analysis, cross-validation–based stability assessment, and evaluation across multiple data splits to ensure robustness and interpretability. Experiments conducted on a real-world high-dimensional clinical dataset demonstrate that the proposed approach achieves consistent predictive performance, with AUC values ranging from 0.628 to 0.641 across five cross-validation folds (mean AUC ≈ 0.633). Performance remains stable under varying train–test splits, indicating strong generalizability. Interaction analysis reveals that a small number of dominant feature interactions—such as age combined with length of hospital stay and medication count combined with diagnostic information—consistently contribute to model predictions, appearing in over 80% of validation folds. Ablation studies further confirm that removing interaction-aware components leads to noticeable performance degradation, highlighting their importance.  In conclusion, this study demonstrates that explicit feature interaction modeling enhances interpretability, stability, and generalization in clinical prediction tasks. The proposed hybrid framework provides a reliable foundation for developing trustworthy and transparent clinical decision-support systems

This research benchmarks multiple machine learning (ML) algorithms for large-scale loan default prediction using a real-world dataset of 255,000 borrower records, where default cases represent only ~9–12% of total observations. The study addresses the persistent gap in comparative analyses of ML models that balance predictive accuracy, interpretability, and computational efficiency for credit risk assessment. Six algorithmic families were evaluated Logistic Regression, Random Forest, XGBoost, LightGBM, CatBoost, Artificial Neural Networks (ANN), and Stacked Ensemble—using standardized preprocessing, hybrid imbalance handling (SMOTE, class weighting, under-sampling), and comprehensive evaluation metrics (AUC, F1, Recall, Precision, PR-AUC, and Brier Score). Empirical results show Logistic Regression achieved the highest AUC of 0.732, outperforming nonlinear models under the baseline configuration, while LightGBM attained perfect recall (1.0) but low precision (0.116), indicating over-prediction of defaults. Gradient boosting models demonstrated robust calibration (Brier ≈ 0.114–0.116) and the best computational efficiency, with LightGBM showing the fastest training and lowest memory use. CatBoost exhibited strong recall but the slowest computation, and ANN underperformed on tabular data (AUC ≈ 0.56). The Stacked Ensemble delivered balanced results with AUC = 0.664 and improved overall stability. These findings confirm that boosting-based models, particularly LightGBM and CatBoost, offer superior scalability and calibration, whereas Logistic Regression remains a valuable interpretable baseline. The study concludes that effective default prediction requires integrating rebalancing, calibration, and threshold optimization to enhance recall and operational deployment reliability in large-scale credit ecosystems.

Physics learning requires higher-order thinking skills, particularly critical thinking and conceptual understanding. However, students’ critical thinking skills and physics learning outcomes at the senior high school level are still relatively low due to teacher-centered instruction and the limited use of innovative learning models and media. This study aimed to examine the effect of the CORE (Connecting, Organizing, Reflecting, Extending) learning model assisted by Lumio by Smart media on students’ critical thinking skills and physics learning outcomes. This research employed a true experimental design with a post-test only control group design conducted at SMAN 4 Jember in the 2025/2026 academic year. The samples consisted of two classes, with class XI Umum 2 as the experimental class and class XI Teknik 3 as the control class. Data were collected through tests and analyzed using the Mann–Whitney U test. The results showed significance values of 0.027 ≤ 0,05 for critical thinking skills which means H0 is rejected and Ha is accepted and 0.020 ≤ 0,05  for physics learning outcomes which means H0 is rejected and Ha is accepted. Therefore, it can be concluded that the CORE learning model assisted by Lumio by Smart media has a significant effect on students’ critical thinking skills and physics learning outcomes.

This study is motivated by the increasing need for accurate modeling and classification of one-dimensional signal data in intelligent systems. The rapid development of deep learning has led to the adoption of more adaptive and complex neural network architectures capable of capturing both temporal dependencies and local patterns in sequential data. This research aims to analyze and compare the performance of several deep learning models, namely Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid Convolutional Neural Network–GRU (CNN–GRU) model for signal data classification. The research method employs a quantitative experimental approach involving data preprocessing, windowing, model training, and performance evaluation. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. The results indicate that the hybrid CNN–GRU model outperforms the other models, particularly in capturing local features and long-term temporal dependencies within signal data. These findings suggest that the integration of convolutional layers and recurrent mechanisms enhances feature representation and learning stability. This study is expected to contribute both theoretically and practically to the development of deep learning models for signal processing and time-series-based intelligent applications.

The integration of Artificial Intelligence (AI) into precision agriculture has significantly improved plant disease recognition; however, many existing deep learning models remain computationally expensive and feature-redundant, limiting their deployment on low-power and edge devices. To address these limitations, this study proposes a lightweight framework for maize leaf disease recognition based on serial deep feature extraction, dimensionality reduction, and machine-learning–based classification. A pre-trained MobileNetV2 network is employed as a fixed feature extractor to obtain discriminative visual representations, while Principal Component Analysis (PCA) is applied to reduce feature dimensionality by approximately 76%, retaining 95% of the original variance and improving computational efficiency. The compressed features are subsequently classified using a Radial Basis Function Support Vector Machine (RBF-SVM), optimized via grid search and cross-validation. Experiments conducted on a four-class maize leaf disease dataset (Northern Leaf Blight, Common Rust, Gray Leaf Spot, and Healthy), with class imbalance handled during training, demonstrate that the proposed MobileNetV2–PCA–SVM pipeline achieves 97.58% accuracy, 96.60% precision, 96.59% recall, and 96.59% F1-score, outperforming the DenseNet201 + Bayesian-optimized SVM baseline (94.60%, 94.40%, 94.40%, and 94.40%, respectively). This improvement corresponds to a 2.98% accuracy gain, a 55% reduction in error rate, an 86% reduction in model parameters (20.31M to 2.75M), and an 85% reduction in model size (81 MB to 12 MB). These results indicate that the proposed framework provides a compact and efficient solution with strong potential for deployment in resource-constrained agricultural environments.

This study explores the integration of deep learning based approaches in real time video content analysis for intelligent human computer interaction (HCI) in multimedia systems. Traditional video analysis techniques, such as rule-based methods and offline processing, struggle with real time performance and adaptability to complex video data. In contrast, the deep learning model used in this research, particularly Convolutional Neural Networks (CNNs), provides high accuracy in object detection, feature extraction, and real time processing. The integration of CNNs with interactive visualization modules enables dynamic adjustments to video content based on user interactions, ensuring a seamless and engaging user experience. The system was benchmarked in terms of its processing speed, accuracy, and responsiveness, showing significant improvements over traditional approaches in real time video analysis. Moreover, the study demonstrates that combining deep learning with real time visualization enhances the efficiency of interactive multimedia applications, making it suitable for dynamic environments such as surveillance, security monitoring, and interactive media. Despite the system's strong performance, challenges such as computational demands in high-resolution video processing were identified, highlighting the need for further optimization. Future work will focus on optimizing the system for different hardware platforms, incorporating multimodal inputs, and refining deep learning models to address computational bottlenecks. This research contributes to advancing HCI by providing insights into the integration of deep learning for real time video content analysis, which is pivotal for enhancing the interactivity and adaptability of intelligent multimedia systems.

Concept drift, the phenomenon where the statistical properties of data streams change over time, poses a significant challenge in machine learning, particularly for long term data streams. Traditional machine learning models, including batch learning and non-adaptive approaches, struggle to detect and adapt to these changes, leading to degraded performance and inaccurate predictions. This study proposes an adaptive computational model designed to detect and respond to concept drift using incremental learning techniques and statistical drift detection mechanisms. The model integrates an Adaptive Drift Detector (ADD) and Incremental Learning System, enabling real-time adjustments to data distribution changes. The model is evaluated across synthetic and real-world datasets, demonstrating its superior ability to detect abrupt, gradual, and recurring drifts compared to traditional models. Experimental results indicate that the adaptive model maintains high prediction accuracy, minimizes false positive rates, and reduces detection delays. Furthermore, the model performs well in resource-constrained environments, making it suitable for real-time applications such as healthcare prediction, fault detection, and IoT systems. Despite its promising performance, the study identifies challenges related to computational complexity and the model’s performance with imbalanced datasets and noisy data. Future research should focus on optimizing the model’s scalability, computational efficiency, and adaptability to more complex data types to ensure broader applicability in dynamic environments. This work contributes to advancing the detection and adaptation of concept drift, offering a robust solution for dynamic and evolving data streams.

Distributed software systems face significant challenges related to data quality due to their complex, decentralized architecture. These systems often involve multiple nodes responsible for processing and storing data, making it difficult to maintain consistency and ensure accurate data across the entire network. In particular, issues like data inconsistency, latency, and data fragmentation are prevalent in distributed environments. To address these challenges, this study proposes an integrated data quality governance strategy that combines real time monitoring and automated anomaly detection using machine learning models. The proposed strategy aims to improve data consistency, enhance anomaly detection capabilities, and reduce the need for manual intervention, ultimately improving overall data governance in distributed systems. Real time monitoring ensures immediate identification of data issues as they occur, while machine learning models, such as autoencoders and Isolation Forests, automate the detection of anomalies based on high reconstruction errors and data isolation techniques. The study evaluates the proposed strategy through real-world distributed system scenarios, comparing its effectiveness to traditional approaches like periodic audits and manual validation. Results demonstrate that the integrated approach leads to faster anomaly detection, reduced data inconsistencies, and improved overall system performance. The use of advanced machine learning techniques and real time analytics significantly enhances the system's ability to maintain high data quality standards across multiple distributed nodes. This strategy has wide-ranging implications for industries that rely on distributed systems, such as finance, healthcare, and IoT, where data integrity is essential for operational success. Future research can focus on integrating more advanced machine learning techniques and optimizing the real time monitoring framework to handle larger and more complex systems.

Explainable Artificial Intelligence (XAI) has become a critical area of research within artificial intelligence, focusing on improving the transparency and interpretability of machine learning (ML) models, often referred to as "black-box" models. The need for XAI techniques arises from the inherent complexity of ML models, which can make their decision-making processes difficult for users to understand. This study investigates various XAI techniques, including LIME (Local Interpretable Model-agnostic Explanations) and SHAP (SHapley Additive exPlanations), to assess their impact on model interpretability without significantly compromising predictive performance. A comparative experimental design was used, applying these XAI methods to different ML models, including deep neural networks and ensemble methods, within large-scale enterprise data analytics systems. The results indicate that XAI methods significantly enhance model transparency and decision traceability, allowing users to understand the influence of individual features on predictions. While a slight reduction in predictive accuracy was observed, especially with simpler models, the trade-off between interpretability and performance was deemed acceptable, particularly in fields requiring transparency, such as healthcare, finance, and autonomous systems. The use of XAI in enterprise data systems has practical implications for fostering trust and enabling informed decision-making among stakeholders. Furthermore, the study discusses the challenges and limitations of applying XAI techniques, such as complexity, scalability, and model-specific limitations. Future research is suggested to focus on developing more scalable and efficient XAI methods, enhancing their applicability across various model types, and addressing the challenges of real-time applications. This will be crucial in ensuring the widespread adoption of XAI in critical domains, promoting the ethical use of AI while maintaining predictive accuracy.

The increasing complexity and scale of modern cybersecurity threats necessitate the development of advanced systems capable of efficiently detecting, analyzing, and mitigating incidents in real time. This paper proposes an automated framework for digital forensics and incident response that leverages big data analytics and real time network traffic profiling. The framework integrates cutting-edge technologies, including Apache Spark for real time data processing and Hadoop for scalable data storage, combined with machine learning models like LSTM and Autoencoders to detect anomalies and threats in network traffic. By automating the process of incident detection and response, this framework significantly reduces the time required to identify threats and improves the accuracy of forensic evidence correlation across heterogeneous network environments. The study highlights the advantages of using machine learning models and big data tools to address the limitations of traditional manual and semi-automated systems, which often struggle to keep pace with large-scale data generation. Testing results demonstrate that the proposed framework can handle large data volumes efficiently, providing real time, actionable insights with significantly reduced response times. Additionally, the framework improves forensic analysis by enabling the correlation of evidence from different devices and protocols, making it more effective than traditional methods in identifying the root cause of security incidents. However, challenges related to data heterogeneity, scalability, and system integration were encountered during testing. The proposed framework holds promise for significantly enhancing the efficiency and effectiveness of cybersecurity operations, with future work focusing on further integration of advanced AI techniques and machine learning models for dynamic and adaptive incident response.

The increasing sophistication of cyber threats has rendered traditional cybersecurity models insufficient in safeguarding enterprise networks. This study introduces a risk aware cybersecurity governance model that integrates real time threat intelligence with predictive anomaly detection to proactively mitigate potential threats. By leveraging advanced machine learning and AI techniques, the model enhances the ability to identify and address cyber threats before they can escalate into significant incidents. The model’s ability to predict anomalies, analyze real time threat intelligence feeds, and provide early warnings allows for faster response times and reduced risk exposure compared to traditional reactive models. Through simulations and real-world use cases, the proposed model demonstrated a 30% reduction in response time and a 25% decrease in overall risk exposure, showing its potential to improve security decision-making and resilience in dynamic threat environments. Unlike traditional models that rely on static rules and periodic policies, the proposed model uses predictive analytics to stay ahead of evolving threats, ensuring continuous monitoring and rapid adaptation. This proactive approach enhances organizational resilience, particularly in handling sophisticated cyber threats such as ransomware, malware, and phishing attacks. Despite its effectiveness, challenges such as data overload, scalability, and the need for interpretability in AI models remain. Future research will focus on refining predictive models, improving scalability for larger networks, and enhancing the explainability of machine learning models to foster greater trust in automated cybersecurity systems. This study contributes to the ongoing evolution of cybersecurity governance by demonstrating the value of integrating predictive and real time monitoring technologies for enhanced threat detection and mitigation.

This study discusses the development of thematic teaching materials in elementary schools to improve the quality of learning in accordance with the 2013 Curriculum. The research method used is research and development with the ADDIE model (Analysis, Design, Development, Implementation, Evaluation). The results show that the thematic teaching materials developed with approaches such as Project Based Learning (PjBL), Problem Based Learning (PBL), and Contextual Teaching and Learning (CTL) can enhance students' critical thinking, creativity, and social responsibility skills. The integration of local wisdom values as well as the use of digital and contextual teaching materials make learning more engaging, interactive, and meaningful. The thematic teaching materials based on these innovative learning models can help students develop 21st-century skills. Thus, the development of thematic teaching materials based on innovative models is an effective strategy for creating quality learning in elementary schools that is relevant to curriculum demands and the needs of today's learners. The use of technology and contextual approaches enriches students' learning experiences, making them more relevant and beneficial in daily life.

In the learning process, educators often face challenges in helping students achieve optimal learning outcomes. One of the contributing factors is the dominance of teacher-centered learning models, which frequently leads to low student engagement and reduced motivation. This condition affects students’ curiosity, particularly in Science learning, which requires active participation and exploration. Therefore, innovative teaching materials are needed to encourage student involvement while simultaneously stimulating their curiosity. One alternative that can be utilized is the Student Worksheet (Lembar Kerja Peserta Didik/LKPD), which can be designed systematically and contextually to support active learning. This study aims to determine the effectiveness of using LKPD in enhancing students’ curiosity in Grade VII Science learning at SMPN 1 Sanrobone. The research employed a Classroom Action Research (CAR) approach carried out in two cycles, each consisting of planning, implementation, observation, and reflection stages. The findings indicate a significant increase in students’ curiosity after the implementation of LKPD, as shown by improvements in questioning behavior, interest in the material, and the ability to observe and explore scientific phenomena. Based on these results, it can be concluded that the use of LKPD is effective in fostering students’ curiosity and contributes to a more interactive and meaningful Science learning experience.

The integration of Artificial Intelligence (AI) in higher education has led to the increasing use of AI-powered adaptive learning models that support personalized and data-driven learning. However, studies examining preservice teachers’ perceptions of these models remain limited, despite their important role in future classroom implementation. This study aims to explore preservice teachers’ perceptions of AI-powered adaptive learning in higher education, focusing on perceived usefulness, learning adaptivity, learning experience, and perceived concerns. A descriptive qualitative research design was employed involving 53 preservice teachers from various universities. Data were collected using a Likert-scale questionnaire and open-ended questions. Quantitative data were analyzed descriptively using percentage distributions, while qualitative data were examined through simple thematic analysis. The findings reveal that preservice teachers generally demonstrate positive perceptions of AIpowered adaptive learning, particularly in terms of learning effectiveness, adaptability, and engagement. Nevertheless, concerns related to over-reliance on AI, ethical issues, and data privacy were also identified. These results indicate that preservice teachers show readiness to engage with AI-supported learning, while highlighting the need for teacher education programs to promote responsible and pedagogically informed AI integration.

The current practice of Islamic Religious Education (PAI) learning in schools still faces the challenge of the dominance of conventional methods that tend to be teacher-centered, so that the aspect of internalization of values by students is often not optimal . This article aims to describe the characteristics and conceptual relationships between the Social Interaction and Personal learning models in the context of PAI, while also examining the relationship between the two models with constructivist and humanistic learning theories . Through a literature study, the analysis shows that the Social Interaction model based on constructivist theory emphasizes the construction of knowledge through group dynamics, collaboration, and social interaction . Meanwhile, the Personal model based on humanistic theory is oriented towards the development of individual potential, self-awareness, and spiritual reflection . The results of the study conclude that the integration of these two models offers a holistic learning approach, where students not only understand PAI material cognitively, but are also able to actualize religious values through real social experiences and deep personal appreciation .

This research aims to determine the effect of learning using the Problem Based Learning model assisted by the Powtoon application on the learning outcomes of Class XI Madrasah Aliyah at the Islamic Education Park (TPI) Sawit Seberang. This research is a type of quantitative research with a Quasy Experimental Design. The research population was 36 students in class XI MIA A and 36 students in class XI MIA B, 36 students in class XI MIA C and 36 students in class, and X MIA C totaling 36 people as the control class. The instrument in this research is in the form of a test in the form of 30 multiple choice questions. Data analysis of student learning outcomes using the t-test formula, also with the help of SPSS version 25 data processing. Data analysis of student learning outcomes with the results of calculating the average learning outcomes shows that the experimental class which uses the Problem Based Learning models is higher than the control class. The prerequisite test is proven that the data is normally distributed and homogeneous. The results of the test calculation were obtained under count worth 9.709 and table worth 1.668 so that it shows tcount > table so Ha2 is accepted. The conclusions in this research explain that there are influences and differences before and after the implementation of the Problem Based Learning model on student learning outcomes in class XI Madrasah Aliyah at the Islamic Education Park (TPI) Sawit Seberang.

The research intends to formulate an integrative approach for designing the Merdeka Curriculum based on Deep Learning in the Economics subject at SMA N 1 Boyolali. The rapid advancement of digital technology demands more adaptive, contextual, and student-centered learning models aligned with 21st- century competencies. Using a an exploratory research approach supported by a structured assessment of prior studies (SLR), this research synthesizes theoretical perspectives and previous studies related to curriculum design, deep learning, and educational innovation. The results reveal that the incorporation of Deep Learning into the Merdeka Curriculum offers a more responsive and data-driven framework, enabling deeper conceptual understanding, critical thinking, and problem-solving skills in Economics learning. However, several challenges were identified, including teachers limited digital competence, insufficient infrastructure, and minimal curriculum innovation training. The study emphasizes the significance of combining technological and pedagogical innovation to strengthen learning quality. The results provide a conceptual model that educators can adopt to design an adaptive curriculum aligned with students’ characteristics and contemporary educational demands.

This study aims to conduct a comparative analysis of curriculum development and teaching methods in Islamic education in Indonesia and Egypt. Both countries have a long tradition in the development of Islamic sciences, but they demonstrate different approaches in accordance with their respective social, cultural, and educational policy contexts. This study uses a literature review method with a descriptive qualitative approach through a review of journals, books, and Islamic education policy documents. The results of the analysis show that the Islamic education system in Indonesia emphasizes the integration of religious and general knowledge in its curriculum, in line with the spirit of religious moderation and the demands of modern society. Meanwhile, Islamic education in Egypt, especially through Al-Azhar University, emphasizes a traditional approach based on tafaqquh fi al-din with a focus on mastery of classical Islamic sciences. In terms of teaching methods, Indonesia is more adaptive to active and technology-based learning models, while Egypt maintains the talaqqi and memorization methods, but has begun to adopt modern pedagogical innovations. Overall, both education systems have their own strengths; Indonesia excels in contextualizing the curriculum, while Egypt sets an example in maintaining the authority and continuity of Islamic scholarship. These findings are expected to contribute to the development of a balanced Islamic education that integrates values and modernity.

The rapid advancement of digital technology has transformed the landscape of education, particularly in secondary schools. This study investigates the influence of digital learning innovation on enhancing deep learning and student engagement. Using a quantitative survey method, data were collected from 150 secondary school students in Indonesia who actively use digital learning platforms such as Learning Management Systems (LMS), mobile learning applications, and hybrid learning models. The research instrument consisted of a closed-ended questionnaire using a five-point Likert scale, which measured students’ perceptions of digital learning innovation, their engagement levels, and indicators of deep learning. Data were analyzed through simple linear regression using SPSS version 25. The results indicate a strong and significant positive relationship between digital learning innovation and both deep learning and student engagement, with a correlation coefficient (R) of 0.732 and an R² value of 0.536. This finding suggests that digital learning innovation explains 53.6% of the variation in students’ deep learning and engagement levels. Students who frequently use digital learning technologies tend to show higher motivation, independence, critical thinking, and reflective learning behavior. The study concludes that digital learning innovation plays a crucial role in supporting meaningful learning and active participation. It highlights the importance of integrating technology strategically into school-based learning to foster adaptive, collaborative, and lifelong learners in the digital era.