Publication Search

Education plays a crucial and essential role in improving the quality of human resource globally, thus a research was conducted with the aim of producing a product, namely a learning module. This module is one of the learning resources that can be used by students independently. Therefore, the module was developed to function as a learniing guide for students. The lerning model used in this research is Creative Problem Solving. In this learning model, the teacher presents problems so that students can find answers innovatively and sharpen  their critical thinking skills. The suitability of this learning module will be evaluated based on three aspect, namely validity, practicality, and effectiveness. Based on the research sample data, this consists of students from class XI Electrical Power Installation Techniques (TITL) 1 at SMKN 1 Driyorejo. This method uses the Research & Development (R&D) approach. The analysis of differences in learning outcomes was carried out using the One Group Pretest-Posttest Method, the treatment in the form of a learning module based on the Creative Problem Solving model was given to student. The research result show that (1) the modules suitability is stated as very valid wth a score of 89,60, (2) aspect, it is stataed as very practical with a total average reaching 90,60, (3) effectiveness is proven from the improvement in learning outcomes in terms of knowledge and domains with an average pretest score of 51,3, while the average posttest score is 85 with a significance of 0.000.

This research aims to develop and assess the feasibility of Microsoft Access-based learning media as a digital archive tool. Using a Research and Development (R&D) approach, the study follows the ADDIE model (Analysis, Design, Development, Implementation, and Evaluation). The participants included one media expert, one material expert, one language expert, and 33 class X MPLB students at SMK Negeri 49 Jakarta. Data collection was conducted through observations, interviews, and questionnaires rated on a Likert scale, evaluated by experts and students. The findings show that: 1) The Microsoft Access-based media, developed for the MPLB basics subject, can be run on laptops or computers; 2) The media suitability received an 81.3% rating ("Very Appropriate"), material suitability 89.3% ("Very Appropriate"), and language suitability 100% ("Very Appropriate"). In the trials, individual testing with three students yielded 89.6% ("Very Eligible"), small group testing with 10 students received 75.7% ("Decent"), and field testing with 20 students reached 86.8% ("Very Eligible"). 3) The product is deemed highly suitable by experts, with strong support from small group and field trials, confirming its effectiveness as a learning tool.

This community service activity was conducted to optimize English language learning through the implementation of the Flipped Classroom model for students of SMP IT Imam Syafi’i 2 Pekanbaru, Riau, Indonesia. The activity aimed to enhance students’ learning engagement, autonomy, and comprehension in English through a student-centered instructional approach. The program was implemented in three stages: preparation, implementation, and evaluation. Learning materials in the form of instructional videos and digital resources were provided prior to classroom meetings to enable students to study independently. Face-to-face sessions were then devoted to interactive discussions, collaborative tasks, and communicative practice to reinforce students’ understanding and language skills. The outcomes of the program indicated increased student participation, improved motivation, and better mastery of English materials. The Flipped Classroom model contributed to creating a more active and meaningful learning environment. Therefore, this community service initiative demonstrates that the Flipped Classroom approach can serve as an effective instructional strategy to support English language learning at the junior high school level

The moral crisis continues to be a concern for Indonesian society today. Therefore, reading literacy in elementary schools is an appropriate strategy that is not only oriented towards the ability to understand texts but also plays an important role in the formation of students' morals and character. The selection of literacy reading materials needs to be adjusted to the stage of moral development of students so that the values contained in the reading can be understood and internalized optimally. This article aims to examine the criteria for literacy reading materials based on the moral development of elementary school students based on Banten Province folklore. The method used is a literature study by analyzing the theory of children's moral development, criteria for selecting literacy reading materials, and moral values contained in Banten folklore. The results of the study indicate that Banten Province folklore has great potential as literacy reading materials that are appropriate to the moral development of  elementary school students because they contain values of honesty, responsibility, hard work, courage, leadership, empathy, social concern, and religiosity. Thus, Banten folklore is relevant to be used as literacy reading materials that support character learning in elementary schools.

This study developed and evaluated an AI-integrated Virtual Reality (VR) system designed to enhance personalized learning in higher education. While VR improves engagement, existing systems often lack adaptivity or experience high latency during AI interactions. To address these limitations, this research introduces a novel integration of a cache-optimized Llama 2 Large Language Model (LLM) that delivers real-time, motivationally grounded feedback. The system was implemented using Unity 3D and validated with 50 undergraduate students. Technical validation showed that the cache layer reduced interaction latency from 17.7 ms to 14.2 ms and maintained zero system crashes throughout the pilot. Learner motivation was assessed using Keller’s ARCS model, yielding mean scores ranging from 4.08 to 4.69 across all dimensions. Independent t-tests (p > 0.05) and negligible effect sizes (Cohen’s d < 0.2) revealed no significant difference between technical (ICT) and non-technical (Physics) students. These findings confirm that the proposed system effectively bridges technological and motivational gaps, providing a robust model for adaptive, immersive education.

This study presents a Graph-Augmented Isolation Forest (GAIF), an unsupervised anomaly-detection framework for analyzing mobile user behavior. The proposed framework represents users and behavioral attributes as a user–feature bipartite graph, enabling the capture of relational dependencies that are not explicitly modeled in conventional vector-based approaches. Low-dimensional user representations are learned through Node2Vec and Graph Sample and Aggregate (GraphSAGE), and the resulting embeddings are subsequently processed by an Isolation Forest to produce anomaly scores. Experiments are conducted on a Mobile Device Usage and User Behavior dataset comprising 700 user profiles derived from application-level behavioral indicators. The dataset is treated as a behavioral abstraction rather than as a malware classification benchmark. A consistent 80:20 stratified train–test split is employed, with all learning-capable operations restricted to the training data to mitigate information leakage. Detection performance is evaluated post hoc using precision, recall, F1-score, and area under the curve (AUC) metrics. Under the evaluated setting, GAIF achieves an F1-score of 0.94 and an AUC of 0.97, demonstrating improved anomaly detection effectiveness relative to representative unsupervised baseline methods. These results are obtained on a static, proxy dataset and should not be interpreted as evidence of real-time deployment capability. Model interpretability is supported through post-hoc Uniform Manifold Approximation and Projection (UMAP) visualizations of the learned embeddings, providing structural insights into anomalous user behavior. Overall, the findings indicate that integrating graph-based representation learning with isolation-based anomaly scoring constitutes a computationally efficient approach for unsupervised mobile user behavior anomaly detection within the scope of this study.

This study presents a comprehensive comparative analysis of four traditional machine learning algorithms Decision Tree, Random Forest, K-Nearest Neighbors, and Support Vector Machine for Android malware detection using the preprocessed TUANDROMD dataset comprising 4,465 instances and 241 features representing both static and dynamic application characteristics. Motivated by the limitations of conventional signature-based and hybrid detection methods, especially in managing imbalanced datasets and detecting emerging malware variants, the study employed SMOTE to ensure balanced training data and fair model evaluation. The dataset was divided into 80% training and 20% testing subsets, and models were assessed using key performance metrics including accuracy, precision, recall, F1-score, and ROC AUC. The findings revealed that the proposed Random Forest model outperformed the other classifiers, achieving an accuracy of 0.993, precision of 0.992, recall of 1.000, F1-score of 0.996, and a near-perfect ROC AUC of 0.9998 surpassing state-of-the-art approaches. These results affirm the superior predictive capability, consistency, and robustness of the Random Forest algorithm in Android malware detection. The study concludes that base models, when integrated with class-balancing techniques, provide reliable and efficient malware detection across imbalanced datasets. For future research, the study recommends exploring advanced hybrid or ensemble frameworks that integrate Random Forest with deep learning architectures or other meta-heuristic optimization techniques to further enhance detection accuracy, adaptability, and resilience against rapidly evolving Android malware threats.

Pneumonia remains a leading cause of morbidity and mortality worldwide, particularly in resource-limited settings and among elderly populations, where timely diagnosis and continuous monitoring are often constrained by limited clinical infrastructure. This study presents an edge–cloud–integrated framework for early pneumonia risk monitoring, leveraging multimodal wearable sensors and deep learning to support continuous short-duration monitoring. The proposed system is designed to operate in near real time under simulated deployment conditions, continuously acquiring and analyzing physiological signals (respiratory rate, heart rate, SpO₂, and body temperature) alongside event-driven acoustic biomarkers (cough sounds) within a distributed architecture. A lightweight edge module performs local signal preprocessing and anomaly triage, selectively transmitting salient information to a cloud-based multimodal deep learning model for refined risk estimation and interpretability analysis. The framework was evaluated using a multi-source dataset comprising public repositories (MIMIC-III and Coswara) and a clinically supervised wearable study conducted in two Nigerian hospitals, resulting in 718  hours of quality-controlled multimodal monitoring data. In a pooled multi-source evaluation, the system achieved an AUC of 0.95, while in a clinically realistic local-only evaluation, the AUC was 0.86, reflecting a consistent but preliminary diagnostic signal. These results highlight the importance of local data adaptation for real-world applicability and suggest that multimodal AI can provide meaningful early risk indicators under resource constraints. Beyond predictive performance, this work demonstrates the feasibility of integrating multimodal learning, edge–cloud computation, and explainable analytics into a deployment-aware, privacy-preserving monitoring framework for low-resource healthcare environments.

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.

The increasing complexity and scale of modern network traffic driven by IoT and cloud-based infrastructures have made accurate intrusion detection a critical challenge. Conventional network intrusion detection systems (NIDS) and many deep learning–based approaches struggle to reliably detect minority and stealthy attacks due to severe class imbalance and limited discrimination of subtle traffic patterns. To address these limitations, this study proposes a hybrid CNN–RBF–Attention framework for network intrusion detection. The proposed model integrates three complementary components: (i) a convolutional neural network for hierarchical feature extraction from network flow data, (ii) a radial basis function (RBF) network for localized nonlinear classification using prototype-based decision regions, and (iii) an attention mechanism that adaptively weights RBF activations to emphasize discriminative traffic patterns. SMOTE is applied exclusively to the training data to mitigate class imbalance. The framework is evaluated on the widely used CICIDS2017 and CICIDS2018 benchmark datasets in both binary and multiclass settings, using recall, precision, F1-score, confusion matrices, and ROC analysis. Experimental results demonstrate that the proposed hybrid model consistently outperforms standalone CNN and RBF baselines, particularly in terms of recall and F1-score. On the CICIDS2018 dataset, the model achieves 99.81% accuracy and 99.81% F1-score in binary classification, and 99.54% accuracy and 99.54% F1-score in multiclass classification. On CICIDS2017, it achieves 98.12% accuracy and 98.12% F1-score in binary classification, and 98.92% accuracy and 98.92% F1-score in multiclass classification. Confusion matrix and ROC analyses further show strong class separability and reliable performance in low–false-positive-rate regions, which is critical for real-world IDS deployment. These results confirm that combining deep hierarchical feature learning, localized prototype-based classification, and attention-guided refinement yields a robust, operationally reliable intrusion detection framework for highly imbalanced network environments.

This article aims to explore the evaluation of the context, inputs, processes, and products of the implementation of Islamic Religious Education and Character Education at SMA Negeri 1 Tikep. Until now, there are still many obstacles faced by a number of educational institutions, especially in the aspects of implementation and evaluation. This study uses a qualitative approach and a type of evaluation research with the CIPP (Context, Input, Process, and Product) evaluation model, which was developed by Stufflebeam. This research was conducted by analyzing the data to answer the problem formulation without testing the hypothesis. The main data from this study were obtained through descriptive analysis, with the data collection process through interviews, documentation, and Post-Test as additional data. The results of this study indicate that from the perspective of the CIPP evaluation model developed by Stufflebeam, the context, input, process, and product aspects of PAI and Budi Pekerti learning based on the 2013 curriculum at SMA Negeri 1 Tikep  are included in the good category.

This study explores the shift from traditional to modernist educational paradigms in Islamic boarding schools (pesantren), which have been in existence since before Indonesia’s independence. The shift involves adapting learning methods while retaining traditionalist approaches, ensuring that they meet modern needs without eliminating their core values. The study uses a qualitative-phenomenological approach to examine three main areas: (1) the strategy of educational values and spiritual practices employed by pesantren leaders, with an emphasis on the TAQWA method, which aims to improve student understanding quickly; (2) the integration of Qur'an literacy, religious traditions, and environmental empowerment in the educational process; and (3) the impact of this model on student character development. Qur'an literacy in this context extends beyond reading and memorizing verses, focusing on understanding and actualizing its values in daily life. Religious practices such as book study, worship routines, and etiquette coaching promote moral development. Environmental activities, such as agriculture and natural resource management, encourage independence and ecological awareness. This holistic approach can serve as a model for character education, blending spiritual, social, and environmental aspects, while strengthening pesantren's role in fostering moral and ecological awareness.

The maritime education sector faces complex challenges in preparing competent seafarers amid the rapid advancement of digital technology. This study investigates the effect of artificial intelligence-based simulations and AI-based competency assessments on competency achievement levels among nautical cadets at Indonesian maritime training institutions. The research design employed a convergent parallel mixed-methods approach, integrating quantitative and qualitative methods to gain a comprehensive understanding. Quantitative data were collected from 150 cadets using a validated questionnaire. In comparison, qualitative data were obtained through in-depth semi-structured interviews with fifteen instructors and ten cadets. Multiple regression analysis revealed that the research model significantly predicted cadet competency achievement. The findings indicate that AI-based assessments exert a stronger influence than AI simulations in improving competency. The qualitative exploration highlighted adaptive feedback mechanisms and personalized learning pathways as critical success factors in implementing learning technologies. This study provides empirical evidence for maritime institutions to prioritize strategic investments in AI-based assessment systems while maintaining a human-centered pedagogy. The research contribution lies in integrating fourth industrial revolution technologies into the training, certification, and watchkeeping standards compliance framework for seafarers, thereby strengthening Indonesia's maritime education ecosystem and aligning it with international standards.

Freshwater fish diseases significantly affect aquaculture productivity and economic sustainability, while accurate visual classification remains challenging due to interclass similarity and image variability. This study presents a comparative evaluation of three deep learning architectures—DenseNet201, ResNet50, and EfficientNetV2-S—using a stepwise optimization strategy combined with Gradient-weighted Class Activation Mapping (Grad-CAM) for freshwater fish disease classification. Models were trained through three phases: baseline, optimized, and fine-tuned. Performance was evaluated using accuracy, precision, recall, F1 score, Matthews correlation coefficient (MCC), Cohen’s kappa, and per-class ROC–AUC. Results show consistent performance improvement across all architectures, with EfficientNetV2-S achieving the highest accuracy (97.14%), followed by ResNet50 (96.11%) and DenseNet201 (94.40%). High ROC–AUC values (>0.98) indicate strong discriminative capability. Grad-CAM analysis confirms that all optimized models focus on biologically relevant lesion regions, enhancing model transparency and reliability.

This study was motivated by the importance of the availability and quality of infrastructure as a major supporting factor for the comfort and smooth running of the lecture process at the postgraduate level. The purpose of this study was to evaluate educational facilities and infrastructure based on the perceptions of postgraduate students in the Islamic Education Study Program (PAI) using the Context, Input, Process, and Product (CIPP) evaluation model. This research is an evaluative study with a qualitative approach, involving PAI graduate students as research subjects. Data collection was conducted through questionnaires, in-depth interviews, and field observations, then analyzed using the Question Discourse technique to obtain a comprehensive understanding of the students' experiences and assessments. The results show that students view facilities and infrastructure as important to very important in supporting the lecture process. In general, the facilities are considered adequate, but there are still limitations in the air conditioning system and internet network stability, which affect the comfort and effectiveness of learning. The implications of this study emphasize the need for continuous improvement in the quality, maintenance, and management of facilities and infrastructure to support the quality of PAI postgraduate education.

This study aims to: (1) Determine the effect of the discovery learning model with the aptitude treatment interaction strategy on the mathematical concept comprehension and self-efficacy of 10th-grade students at SMA Negeri 3 Tarakan; (2) Determine the effect of the discovery learning model with the aptitude treatment interaction strategy on the mathematical concept comprehension of 10th-grade students at SMA Negeri 3 Tarakan; (3) Determine the effect of the discovery learning model with the aptitude treatment interaction strategy on the self-efficacy of 10th-grade students at SMA Negeri 3 Tarakan. The research population included all tenth-grade students of SMA Negeri 3 Tarakan in the 2024/2025 academic year. The research sample consisted of two classes selected randomly: one experimental class receiving discovery learning with an aptitude treatment interaction strategy and one control class receiving conventional learning. The research instruments consisted of a test measuring mathematical concept understanding on trigonometry material and a self-efficacy questionnaire. The data obtained were tested for prerequisites through normality and homogeneity tests before being analyzed using inferential statistical tests in the form of an independent samples t-test with the assistance of SPSS software version 26.0. The research results show that the implementation of the discovery learning model with the aptitude-treatment interaction strategy has a significant impact on students' mathematical concept understanding and self-efficacy simultaneously, with a significance value of 0.006 < 0.05. Partially, this learning model has a significant effect on students' mathematical concept understanding, with a significance value of 0.018 < 0.05. However, the effect of the discovery learning model with the aptitude-treatment interaction strategy on students' self-efficacy is not statistically significant, as indicated by a significance value of 0.089 > 0.05, even though there is a tendency for increased self-efficacy among students participating in the experimental class learning. Nevertheless, the influence of the discovery learning model with the aptitude treatment interaction strategy on students' self-efficacy is not statistically significant in partial terms, although there is a tendency for an increase in self-efficacy among students participating in the experimental class. These findings suggest that the discovery learning model with the aptitude treatment interaction strategy is effective in improving students' understanding of mathematical concepts in trigonometry material and has the potential to support the development of self-efficacy in mathematics learning.

Islamic boarding schools (Islamic boarding schools) are required to integrate classical scholarly traditions and modern education, one way of doing this is through the integration of the study of yellow books (tahfidzul Qur'an) and Qur'an memorization (tahfidzul Qur'an). The Baitul Abidin Darussalam Wonosobo Tahfidzul Qur'an Islamic Boarding School implements an integrative learning system to balance Qur'an memorization and understanding of Islamic law (shari'a). However, it still faces obstacles such as a tight schedule, different methods, and weak coordination and evaluation. This study examines the implementation patterns, challenges, and impacts of this system, with the hope of serving as a reference for developing a balanced and sustainable model of Islamic boarding school education. This research uses a qualitative approach with a case study design to understand in-depth the implementation of the integrative learning system between Qur'an memorization and the study of yellow books (tahfidzul Qur'an) at the Baitul Abidin Darussalam Wonosobo Islamic Boarding School (PPTQ). Subjects were selected purposively, including the boarding school administrator, tahfidz teachers, yellow book teachers, and students. Data were collected through in-depth interviews, participant observation, and documentation studies. Data analysis was conducted interactively using the Miles and Huberman model, which encompasses data reduction, data presentation, and conclusion drawing and verification to obtain a holistic and contextual understanding. The discussion shows that the integrative learning system in Islamic boarding schools is implemented through a balanced daily schedule between Quran memorization and yellow book study, allowing memorization, understanding, and moral development to occur simultaneously within the students' daily routines. Integration is achieved structurally through scheduling, methodologically by linking verse memorization with book study, and culturally through the instillation of values, etiquette, and pesantren traditions. The success of integration is supported by the exemplary behavior of the kiai (Islamic teachers) and ustadz (Islamic teachers), the religious environment, and the motivation and discipline of the students, despite challenges such as busy schedules, physical exhaustion, differences in student abilities, and limited facilities. The impact of implementing this system is seen in the improved quality of contextual memorization, a more critical understanding of the scriptures, the formation of disciplined and moral character, and the holistic spiritual development of students.

This research develops a machine learning model to classify customer loyalty using the Random Forest algorithm. Customer churn is a critical issue that reduces revenue and increases acquisition costs. A dataset of 50,000 customers from global e-commerce and subscription platforms was processed through data cleaning, imputation, outlier handling, and class balancing with SMOTE. The Random Forest model was built as a baseline and optimized with hyperparameter tuning. Evaluation using accuracy, precision, recall, and F1-score shows that the optimized model achieved 90.81% accuracy and 83.87% F1-score, outperforming previous Naïve Bayes approaches. Feature importance analysis highlights customer service interactions, lifetime value, and demographic factors as key predictors of churn. These findings demonstrate Random Forest’s effectiveness in churn prediction and provide practical insights for customer retention strategies

This study aims to develop a pop-up book learning medium with the theme “Einkaufen” (Shopping) as a German language instructional medium for Grade XI students at SMA Negeri 3 Binjai. The background of this study is the low level of students’ understanding of the learning materials and their lack of learning motivation due to the use of monotonous conventional media. This research employs a Research and Development (R&D) approach using the ADDIE development model, which includes the stages of Analysis, Design, Development, Implementation, and Evaluation. The results of the implementation stage indicate that 98.3% of students agreed that the pop-up book is effective for use in German language learning, and 96.7% stated that the material content is easy to understand. The pop-up book consists of six pages containing learning materials and exercises that focus on reading, writing, and speaking skills. Validation by material and media experts yielded a score of 90, categorized as very good. These findings demonstrate that the developed pop-up book medium is not only visually engaging but also capable of enhancing students’ motivation and comprehension in an integrative German language learning process. This medium can serve as an effective alternative for improving the quality of foreign language learning at the secondary school level.

Open-pit mining activities in Indonesia contribute significantly to the national economy but require stringent monitoring to mitigate environmental degradation. Conventional monitoring methods relying on terrestrial surveys are often constrained by vast coverage areas, high operational costs, and limited field accessibility. This study aims to develop an artificial intelligence model capable of automatically detecting and mapping mining areas to enhance surveillance efficiency. The applied method is Deep Semantic Segmentation utilizing the U-Net Convolutional Neural Network (CNN) architecture. The model was trained using Sentinel-2 satellite imagery, focusing exclusively on Red, Green, and Blue (RGB) spectral channels to replicate human visual perception. Experimental results demonstrate that the proposed model performs reliable segmentation of mining areas, achieving an Accuracy of 93.58% and a Global Intersection over Union (IoU) of 0.8067. These findings indicate that the U-Net architecture can effectively extract spatial features of mines even when utilizing standard visual data. This research contributes to the development of an efficient, cost-effective, and scalable digital monitoring prototype to support innovation in sustainable environmental governance.