Publication Search

Arabic grammar (nahwu) instruction has long been dominated by rule-based approaches that emphasize memorization and formal analysis, often resulting in rigid learning structures and limited responsiveness to learners’ cognitive diversity. While such approaches play an important role in preserving grammatical accuracy, they frequently overlook individual learning trajectories, cognitive readiness, and adaptive instructional needs. In the era of artificial intelligence (AI), language education is increasingly shaped by adaptive learning systems that personalize content, pacing, and instructional strategies based on learners’ profiles. This study aims to reconceptualize Arabic grammar instruction by proposing a conceptual framework that integrates traditional nahwu principles with adaptive learning systems informed by AI. Using a qualitative conceptual analysis, this paper synthesizes classical Arabic grammar pedagogy, contemporary theories of adaptive learning, and recent developments in AI-supported language instruction. The proposed framework highlights key components, including learner profiling, cognitive-level alignment, hierarchical nahwu content structuring, and AI-assisted scaffolding mechanisms. The findings suggest that adaptive learning systems offer significant pedagogical potential to transform nahwu instruction from a static, rule-centered model into a flexible, learner-centered process. This reconceptualization is expected to enhance grammatical comprehension, reduce cognitive overload, and promote learner autonomy in Arabic language education, particularly in Islamic higher education contexts. The study concludes by discussing pedagogical implications and directions for future empirical research on AI-assisted Arabic grammar learning.

Public health is a crucial aspect of improving well-being, particularly at the village level. However, in Srigading Village, Lawang District, Malang Regency, access to information related to health services remains a significant challenge, resulting in suboptimal utilization of Posyandu and Puskesmas services. To address this issue, a web application called “Srigading Sehat” has been developed as a digital platform providing health service information through interactive maps and service schedules. This community service activity employs the IDEA approach (Identify Objective, Design Action Steps, Engage Our Plan, Assess and Follow Up) as a systematic framework. The Identify Objective stage identified limitations in health information access; Design Action Steps designed a participatory learning-based application development model; Engage Our Plan implemented application development, socialization, and training for 25 participants comprising village officials, posyandu cadres, and the general community; and Assess and Follow Up evaluated achievements through Firebase Analytics monitoring. Results showed increased health information access marked by 500 health information page visits within the first 30 days of launch, with total Firebase reads reaching 631 (+1,026.8%). Four village health facilities are actively registered and using the application to independently update their service information. This application is anticipated to serve as an appropriate technological innovation contributing to the enhancement of health services at the village level.

Flower disease detection is a significant challenge in modern agriculture, particularly with factors such as changes in leaf color, petal shape and structure, and environmental conditions affecting the accuracy of conventional models. These factors make it difficult to achieve optimal results using traditional methods. Transfer learning is an effective solution to improve image detection performance, especially when data is limited. This study used several pre-trained models, namely VGG16, ResNet50, and EfficientNet-B0, to detect three types of flower diseases: black spot on roses, white powdery mildew, and leaf rust. The research process included data processing, increasing the data volume using augmentation techniques, model training, and evaluation of the results. Experimental results showed that the EfficientNet-B0 model produced the highest accuracy of 97.2%, significantly better than the CNN model built from scratch with an accuracy of 85.1%. This study demonstrates that transfer learning is highly effective in improving the accuracy of flower disease detection, making it a more reliable alternative to methods that do not utilize pre-trained models, especially for agricultural applications that require high levels of accuracy in disease detection.

Sentiment analysis is a technique in Natural Language Processing (NLP) used to identify user opinion tendencies based on textual reviews. This study analyzer user reviews of the Maxim application on the Google Play Store and compares three Machine Learning algoritmhs-Naïve Bayes, Support Vector Machine (SVM), and CatBoost-in classifying sentiment. The research stages include data collection, text preprocessing, feature extraction using TF-IDF and Chi-Square, class balancing using SMOTE, and performance evaluation through Accuracy, Precision, Recall, and F1-Score. ANOVA is used to examine the influence of feature selection on model performance. The results show that each model exhibits different performance level across the tested feature combinations. The CatBoost achieved the highest accuracy of 99,26% and demonstrating the most stable performance. Meanwhile, the Naïve Bayes and SVM models experienced performance decreases experiments, especially after applying SMOTE. These findings indicate that the choise of algorithm, feature extraction method, and class balancing technique significantly affects classification outcomes. Overall, CatBoost is identified as the best-performing model, providing more consistenst classification result in accordance with the characteristics of the user reviews.

Islamic character education faces dual challenges: declining moral values among students and conventional teaching methods inadequate for developing higher-order thinking skills. This study examines the implementation of a smart learning environment (SLE) model to enhance metacognitive understanding in Aqidah Akhlak (Islamic Creed and Ethics) instruction among eighth-grade students at MTs Salafiyah Merakurak. Employing a mixed-methods action research design, involved 16 eighth-grade students divided into two groups. Data collection utilized classroom observations, semi-structured interviews, and learning documentation. The SLE model integrated Google Classroom, interactive video content, WhatsApp Group discussions, and Google Forms assessments to create a technology-enhanced learning ecosystem. Findings revealed substantial improvements in students' metacognitive capacities: planning skills increased from 25% to 75%, monitoring abilities rose from 31% to 81%, and evaluation competencies grew from 19% to 69%. Students demonstrated enhanced learning autonomy, active participation in collaborative discussions, and improved self-reflection on content comprehension. The SLE approach successfully fostered engaging learning experiences while facilitating deeper internalization of Islamic ethical values. However, implementation encountered constraints including limited technological infrastructure and varied digital literacy levels among students. This research underscores the critical need for developing teachers' digital competencies and strengthening madrasah technological infrastructure to optimize technology-integrated Islamic education.

This study aims to explain the application of various differentiated learning techniques based on Problem-Based Learning (PBL) to improve analytical and writing skills related to procedural texts among junior high school students. This research is based on students' lack of ability to understand and compose procedural texts methodically and in accordance with language conventions, which is caused by the prevalence of conventional teacher-centered learning. This study used a descriptive qualitative methodology involving seventh-grade students from a junior high school that has adopted the PBL model in Indonesian language subjects. Data were collected through observation, interviews, and documentation, then analyzed qualitatively. The results showed that the application of PBL along with differentiated learning and TPACK increased student engagement, accommodated diverse learning styles, and fostered critical thinking, analytical abilities, and collaborative skills. Learning became more meaningful and relevant, enabling students to compose procedural texts more effectively. This study recommends the application of the PBL model with differentiation as an innovative strategy to improve the quality of Indonesian language education in junior high schools.

The rapid development of artificial intelligence, particularly ChatGPT, has created new opportunities to support students’ academic activities in higher education. However, its utilization needs to be evaluated in terms of the alignment between academic task characteristics and technological capabilities to ensure optimal outcomes. This study aims to examine the feasibility of using ChatGPT in students’ academic activities by applying the Task–Technology Fit (TTF) model. This research employed a quantitative approach using Structural Equation Modeling based on Partial Least Squares (SEM-PLS). Data were collected through questionnaires distributed to university students and analyzed using SmartPLS 4 software. The variables examined included Task Characteristics, Technology Characteristics, Task–Technology Fit, Performance Impact, and Utilization. The results indicate that Task Characteristics and Technology Characteristics have a positive and significant effect on Task–Technology Fit. Furthermore, Task–Technology Fit significantly influences Performance Impact and Utilization. Performance Impact also shows a positive and significant effect on the utilization of ChatGPT by students. These findings suggest that the alignment between academic task requirements and the capabilities of ChatGPT plays a crucial role in improving students’ performance and encouraging sustained technology use. The implications of this study highlight the importance of selective and purposeful use of ChatGPT in higher education and provide a reference for higher education institutions in formulating policies related to the ethical and effective integration of artificial intelligence technologies as learning support tools.

This article develops a conceptual model that explains how social capital and digital literacy interact in shaping Islamic financial literacy in the digital era. Through a comprehensive literature review, this study synthesizes theories, empirical findings, and thematic patterns derived from reputable academic journals, scholarly books, and institutional publications. The analysis shows that social capital functions as a value foundation encompassing trust, collective norms, and behavioral orientations that influence individuals’ initial acceptance of sharia-based financial practices. Information obtained through family, religious communities, and social networks becomes a crucial entry point that shapes early perceptions and preferences toward Islamic financial products. Meanwhile, digital literacy strengthens individuals’ ability to access, evaluate, and verify Islamic financial information independently through various digital content such as online articles, infographics, educational videos, and Islamic fintech platforms. The interaction between these two dimensions creates a layered learning process in which social capital provides contextual value and trust, while digital literacy deepens technical understanding in a more objective manner. This article contributes theoretically by proposing the Social Capital–Digital Literacy Integrative Model and offers practical implications for Islamic financial institutions, regulators, and fintech providers in designing more effective strategies to enhance Islamic financial literacy in society.

This study examines the role of mathematics learning in improving logical thinking skills in early childhood. The background of this study is based on the importance of logical thinking skills as a foundation for children's cognitive development, which can begin at an early age through appropriate mathematics learning. The purpose of this study is to analyze how mathematics learning can stimulate the development of logical thinking in early childhood and explore effective learning strategies. The methods used are library research and observation of several models of mathematics learning for early childhood in early childhood education institutions. The findings indicate that fun and concrete activity-based mathematics learning can improve children's abilities in critical thinking, constructing patterns, drawing conclusions, and solving simple problems. The implications of this study emphasize the need for the application of creative and interactive mathematics learning methods to support the development of logical thinking from an early age, while also encouraging educators to integrate mathematics into children's daily activities. This study also recommends the development of learning media appropriate to children's developmental stages for optimal results.

This study evaluates the AIESEC Incoming Global Volunteer (IGV) Program at the Veteran National Development University in Yogyakarta using the CIPP (Context, Input, Process, Product) evaluation model. Employing a descriptive qualitative approach, data were collected through interviews, non-participatory observation, and documentation studies, with validity ensured through triangulation. The findings reveal that the IGV Program is highly relevant to the university’s internationalization agenda and contributes significantly to strengthening cross-cultural competencies among students. The availability of resources and the overall implementation of the program have been effective, though improvements are needed in ensuring consistent mentoring for international participants. The evaluation highlights that the program has generated positive outcomes, particularly in enhancing intercultural competencies and fostering collaboration with local partners. These results underscore the importance of sustaining and refining the IGV Program as a strategic initiative to support global engagement and student development.

Magnetic resonance imaging (MRI) provides rich anatomical contrast for brain tumor assessment, yet routine interpretation remains time-intensive and demands high precision. This work develops a pipeline for four-class brain MRI image classification (glioma, meningioma, pituitary tumor, and no tumor) by combining automated brain-region cropping, data augmentation, and transfer learning with EfficientNetB1. Experimental results demonstrate exceptional performance, achieving an overall accuracy of 0.99 (99%) on the test set. Specifically, the model reached an F1-score of 1.00 for the no tumor class, 0.99 for pituitary, and 0.98 for both glioma and meningioma classes. Beyond reporting numerical performance, the study utilizes Grad-CAM heatmaps to verify that predictions rely on clinically plausible regions rather than spurious background cues. These results indicate that an efficiency-oriented backbone, paired with systematic preprocessing, can achieve reliable and interpretable performance for brain tumor classification tasks.

The increasing social and cultural diversity of contemporary societies has positioned schools as strategic spaces for nurturing inclusive attitudes and managing the challenges of pluralism. Many schools still face obstacles such as limited teacher readiness, inadequate curriculum representation of diversity, and school cultures that are not fully supportive of inclusive practices. This study aims to analyze school strategies in fostering students’ inclusive dispositions in response to pluralistic challenges. The research employed a descriptive qualitative approach through library research, using purposive sampling to select relevant scientific literature. Data were analyzed using an interactive model that involved data reduction, data display, and conclusion drawing. The findings indicate that participatory learning strategies, the application of Universal Design for Learning, and project-based learning are effective in strengthening students’ inclusive attitudes, empathy, and tolerance. The role of teachers as role models, the development of inclusive school culture, and active community involvement were identified as key supporting factors for successful inclusive education. The implications of this study highlight the importance of synergy between teacher professional development, curriculum adaptation, and school policy reinforcement to establish equitable, inclusive, and sustainable educational practices in pluralistic contexts.

This study develops an interpretable machine learning model to predict the settlement status of Hajj fees in Jambi Province, Indonesia. Utilizing the XGBoost algorithm on a dataset of 4,332 prospective pilgrims from 2025, the research addresses the critical challenge of class imbalance where only 28.5% of samples are labeled "Unsettled". The baseline XGBoost model achieved a ROC-AUC of 0.7778, with a recall of 0.3482 for the minority class. SHAP (SHapley Additive exPlanations) analysis was employed to interpret model predictions, revealing that financial features specifically NILAI_VA (Virtual Account Value), JML_SETORAN (Deposit Amount), and JML_PELUNASAN (Settlement Amount) are the most significant factors influencing repayment risk, with negative SHAP values indicating increased default probability. The findings demonstrate that an interpretable XGBoost framework can provide both predictive accuracy and actionable insights for policymakers, enabling targeted interventions such as flexible payment schemes and enhanced financial monitoring for high-risk pilgrims..

The rapid development of digital streaming platforms such as Netflix has generated a large volume of content data with diverse characteristics, thereby requiring effective analytical methods to understand emerging patterns and trends. This study aims to classify Netflix content into two main categories, namely movies and television shows, and to analyze genre trends and content characteristics using a data mining approach with the Naive Bayes algorithm. The dataset used in this study is the Netflix Shows dataset, consisting of 8,809 content entries, with the primary features analyzed including genre, rating, and country of production. The research process begins with data exploration and preprocessing stages, including data cleaning, handling missing values, and transforming categorical features to enable effective model construction. Subsequently, the dataset is divided into training and testing sets to objectively and systematically build and evaluate the Naive Bayes classification model. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics to assess the model’s ability to accurately distinguish between Netflix content types. The experimental results demonstrate that the Naive Bayes algorithm is able to classify Netflix content into Movie and TV Show categories with accuracy, precision, recall, and F1-score values of 100%, respectively. The confusion matrix indicates that no misclassification occurred, suggesting that genre, rating, and country of production features provide a very clear separation between content classes. These findings indicate that the Naive Bayes algorithm can achieve exceptionally high classification performance with optimal evaluation results. The results further reveal distinct differences in characteristics between movies and television shows based on genre and production attributes. Therefore, this study is expected to contribute to the development of content recommendation systems and strategic content management within the streaming industry.

AImbalanced data remains a significant issue in heart disease classification using machine learning, as it tends to cause models to overestimate the majority class while ignoring minority classes with high clinical value. This can lead to a decrease in accuracy and the model's ability to accurately detect disease cases. Therefore, this study aims to assess the effectiveness of oversampling techniques, namely Random Oversampling and Synthetic Minority Oversampling Technique (SMOTE), in improving the performance of the K-Nearest Neighbors (KNN), Naive Bayes (NB), and Random Forest (RF) algorithms. The dataset used comes from Kaggle and consists of 918 data sets with 12 attributes representing patient information related to heart disease prediction. The research stages include data preprocessing, baseline model testing, and re-evaluation using the two oversampling methods. Experimental results show that oversampling can improve the performance of all algorithms. KNN achieved the best results with SMOTE, with an accuracy of 72.98% and an F1-score of 75.39%. In the Naive Bayes algorithm, both oversampling techniques produced relatively stable performance, with the highest F1-score of 73.56% using SMOTE. Meanwhile, Random Forest showed the most optimal performance when combined with Random Oversampling, with an accuracy of 79.19% and an F1-score of 81.51%. These findings confirm that the success of data balancing techniques is strongly influenced by the characteristics of the classification algorithm used, and provide a practical contribution in determining strategies for handling imbalanced data in health research.

Student dropout remains a significant challenge for higher education institutions as it impacts academic quality, educational management efficiency, and students' success in completing their studies. Therefore, an approach that can identify students at risk of dropping out is necessary so that timely academic interventions can be made. This study aims to develop a dropout detection model using an Artificial Neural Network (ANN). The data used come from a publicly available higher education dataset, ensuring research reproducibility. Data preprocessing steps were carried out to improve data quality before modeling, and the Synthetic Minority Over-Sampling Technique combined with Edited Nearest Neighbors (SMOTE-ENN) was applied to address class imbalance issues. The ANN model's performance was evaluated using accuracy, precision, recall, F1-score, and area under the ROC curve (ROC-AUC). The test results show that the ANN model can provide excellent predictive performance in detecting at-risk students. The application of SMOTE-ENN also proved to enhance the model’s sensitivity toward the minority class, as indicated by improvements in recall and F1-score. These findings indicate that the developed ANN model has the potential to be used as a student dropout detection system to support data-driven decision-making and strategy development within higher education institutions.

This study aims to describe the integration of environmental education into early childhood learning activities at TK Prawidya Dharma Demulih through the use of recycled waste as a creative and educational learning medium. The study was motivated by the low environmental awareness among children and the limited use of environmentally themed learning media in the institution. A qualitative descriptive approach was applied using the ADDIE development model, consisting of the stages of analysis, design, development, implementation, and evaluation. The findings reveal that employing a recycled-material spinner game enhanced children’s understanding of environmental cleanliness and encouraged environmentally responsible behavior through playful learning activities. The children showed strong enthusiasm, participated actively, and began to develop habits related to cleanliness after the learning sessions. Moreover, teachers gained new insights into designing innovative and functional learning media using discarded materials. Overall, the use of recycled waste as an educational tool proved effective in fostering environmental awareness while supporting creativity and meaningful learning experiences for early childhood learners..

This community service activity aims to strengthen the role of higher education institutions in preserving local culture through the revitalization of cultural arts learning based on local wisdom, particularly traditional Nias carving art. The main problems faced by the partners include the limited availability of contextual cultural arts learning, minimal integration of traditional art practices into university courses, and students’ low understanding of the philosophical values embedded in Nias carving motifs. The implementation method employs a participatory–educational approach consisting of preparation, socialization, theoretical and practical training, intensive mentoring, and evaluation stages. This activity involves students and lecturers as participants as well as agents of cultural preservation. The results indicate a significant improvement in participants’ knowledge of the symbolic meanings of Nias carving motifs, their basic skills in designing and drawing carving motifs, and their appreciative attitudes toward the preservation of local cultural arts. This activity contributes to the strengthening of cultural arts learning in higher education and has the potential to serve as a sustainable model of community service based on local culture.

This study aims to classify customer payment methods at 17 Coffee & Eatery using machine learning algorithms, namely Naïve Bayes and Support Vector Machine (SVM). The increasing use of digital and non-cash payments has generated large volumes of transaction data that are rarely analyzed optimally, even though such data contain valuable information for business decision making. This research used secondary transaction data collected from January to March 2025, consisting of 10,147 transaction records. The dataset included several attributes such as order time, payment time, transaction type, total sales, number of items, and payment method. Data preprocessing was performed through data cleaning, feature engineering, normalization, and label encoding before being divided into training and testing sets with an 80:20 ratio. The Naïve Bayes and SVM models were then trained and evaluated using accuracy, precision, recall, F1-score, and ROC–AUC metrics. The results show that both algorithms were able to classify payment methods effectively, but SVM achieved higher accuracy and more stable performance than Naïve Bayes. These findings indicate that SVM is more suitable for handling complex and heterogeneous transaction patterns. The implementation of machine learning for transaction classification can support more efficient financial management and data-driven decision making for small and medium enterprises in the culinary sector.

Sudden Infant Death Syndrome (SIDS) is a sudden and unexpected death in infants that is often associated with the prone sleeping position. This study aims to develop an automated monitoring system capable of detecting SIDS risk factors using the YOLOv8 algorithm and to analyze the effect of data augmentation on model performance. The dataset consists of two classes, baby-lying-on-back (supine) and baby-lying-on-stomach (prone), which were processed through model training and evaluation using precision, recall, F1-score, and mAP metrics. The model was trained under two scenarios, without data augmentation and with data augmentation. The results show that the model without augmentation achieved a precision of 90%, recall of 85%, F1-score of 86%, and mAP50 of 93.7%. After applying augmentation, performance improved to a precision of 90%, recall of 87%, F1-score of 88%, and mAP50 of 95.1%. These findings indicate that augmentation increases detection accuracy and enhances model generalization, including robustness against variations in lighting and camera angles. Furthermore, testing with image and video inputs revealed that the non-augmented model exhibited a tendency toward overfitting, particularly in favor of the baby-lying-on-stomach, whereas the augmented model successfully classified both classes accurately. The developed system is also equipped with an alarm feature and early-warning notifications via Telegram to smartphone when a prone position is detected for a certain duration. Overall, the results demonstrate that YOLOv8 with data augmentation is effective for an automated, non-invasive monitoring system for infants, making it suitable for detecting and preventing potential SIDS risk factors.