Publication Search

Polusi udara partikulat halus (PM2,5) merupakan ancaman serius bagi kesehatan masyarakat di Kabupaten Brebes, Jawa Tengah. Faktor penyumbang utamanya adalah emisi kendaraan di jalur Pantura, aktivitas industri perikanan, serta konsentrasi tinggi selama musim kemarau (Juni–November). Tidak adanya model peramalan sub-jam yang akurat menghambat pengembangan sistem peringatan dini yang efektif. Penelitian ini mengembangkan dan mengevaluasi model deep learning berbasis Transformer untuk memprediksi konsentrasi PM2,5 dengan resolusi waktu 15 menit. Data yang digunakan berasal dari NASA GEOS-CF (band PM25_RH35_GCC) yang diakses melalui Google Earth Engine menggunakan API Python. Dataset mencakup periode 1 Januari hingga 22 November 2025, menghasilkan 7.813 observasi per jam, yang kemudian diinterpolasi linear menjadi 31.249 titik data dengan resolusi 15 menit. Arsitektur Transformer terdiri dari 3 lapis enkoder, 4 kepala perhatian multi-head, dimensi embedding 128, dimensi feed-forward 256, panjang sekuen 60 timestep, dan augmentasi fitur menggunakan rerata bergulir (*rolling mean*, jendela = 3) dan beda pertama (*first difference*). Pelatihan dilakukan dengan TensorFlow-Keras, pengoptimal Adam, penjadwal peluruhan kosinus (*cosine decay scheduler*), dan fungsi kerugian Huber. Pembagian data dilakukan secara kronologis: 70% pelatihan, 30% validasi. Evaluasi pada set uji independen (16 Agustus–21 November 2025, 9.357 observasi atau 97 hari 11 jam 15 menit) menghasilkan MAE 0,7691 µg/m³, RMSE 1,2052 µg/m³, R² 0,9945, dan *Explained Variance Score* 0,9948. Model ini mampu menggambarkan variasi diurnal dan anomali musiman secara akurat, jauh melampaui model LSTM dan GTWR konvensional. Penelitian ini memberikan kontribusi signifikan di bidang Teknologi Informasi melalui kerangka kerja pengolahan *big data* satelit untuk aplikasi lingkungan.

Penyakit jantung tetap menjadi penyebab utama kematian di seluruh dunia, sehingga menekankan perlunya deteksi dini dan langkah pencegahan aktif melalui pendekatan yang didasarkan pada analisis data. Penelitian ini bertujuan untuk memprediksi kemungkinan terjadinya penyakit jantung dengan memanfaatkan berbagai metode pembelajaran mesin, yang bergantung pada indikator kesehatan seperti nilai tekanan darah, tingkat kolesterol, indeks massa tubuh, serta pengukuran denyut nadi. Penelitian ini menerapkan teknik pembelajaran terawasi, meliputi Regresi Logistik, Pohon Keputusan, Hutan Acak, dan Mesin Vektor Pendukung, untuk mengevaluasi efektivitas masing-masing model dalam menentukan kondisi kesehatan pasien. Data yang digunakan berasal dari repositori kesehatan yang dapat diakses secara gratis, mencakup 303 catatan medis pasien, di mana setiap catatan ditandai oleh 14 atribut kesehatan yang berbeda. Untuk memastikan keandalan data, beberapa langkah pra-pemrosesan diterapkan, seperti normalisasi, seleksi fitur, dan penanganan data yang hilang. Temuan eksperimen mengungkapkan bahwa algoritma Hutan Acak menunjukkan akurasi tertinggi sebesar 92,3%, mengungguli model lainnya dalam aspek presisi, recall, dan skor F1. Hal ini menandakan bahwa metode Hutan Acak unggul dalam mendeteksi pola dan faktor yang memperbesar risiko penyakit jantung. Penelitian ini berkontribusi pada pengembangan sistem diagnostik cerdas, yang dapat mendukung tenaga medis dalam pengambilan keputusan awal yang didasarkan pada bukti kuat. Kesimpulan dari studi ini menyoroti pentingnya memasukkan teknologi pembelajaran mesin ke dalam ekosistem layanan kesehatan digital guna meningkatkan ketepatan diagnosis dan memperbaiki hasil kesehatan pasien.

This study aims to explore the effectiveness of an integrated Islamic learning model in optimizing intellectual (IQ), emotional (EQ), and spiritual (SQ) intelligence in early childhood at an integrated Islamic Early Childhood Education (PAUD) in Kotabumi. The study used a qualitative method with a case study approach. Subjects included teachers, parents, and children aged 4–6 years, while data were collected through participant observation, in-depth interviews, and learning documentation. The results indicate that the integrated Islamic learning model positively influences the development of children's IQ, EQ, and SQ. Activities such as reading the Hijaiyah alphabet, Islamic logic games, and Quranic stories enhance children's cognitive skills. Activities based on Islamic values, such as moral stories, empathy exercises, sharing, and group work, strengthen emotional intelligence, while prayer, Surah memorization, and the Prophet's example foster spiritual intelligence. The holistic integration of these three dimensions of intelligence creates intelligent, emotionally mature, and spiritually strong children.

This community engagement program aims to strengthen the economic capacity of community groups in Azerbaijan through an integrated model combining artificial intelligence (AI)–supported learning, digital skills enhancement, and practical economic training. Although Azerbaijan has demonstrated rapid digital progress, micro-entrepreneurs still face challenges related to limited digital literacy, insufficient financial planning capabilities, and the absence of structured mentorship that integrates emerging technologies. This program utilizes a participatory development approach involving baseline assessment, AI-driven training modules, intensive workshops, and mentoring over two months. A total of 45 participants from three community clusters were included. Quantitative findings reveal a 63% increase in digital literacy scores and a 67% adoption rate of AI-supported tools for marketing and financial planning. Qualitative insights show improved confidence in business decision-making, better understanding of market dynamics, and enhanced practical skills in budgeting and content creation. Participants also reported early indicators of improved sales performance and customer engagement. The integration of AI tools into community-based capacity-building initiatives demonstrates high effectiveness in accelerating digital readiness and entrepreneurial resilience. The program concludes that structured AI-based interventions can significantly empower grassroots communities in emerging economies. Limitations include the small sample size and short mentoring duration. Future research should explore larger populations and long-term economic outcomes.

The purpose of this study is to discuss the introduction of the Pendet Dance in the Markandeya–Walailak cultural exchange program and to evaluate its effectiveness as a cross-cultural character education tool. To investigate program design, execution, and results, the study used a descriptive qualitative approach backed by the CIPP (Context, Input, Process, Product) evaluation model. During the cultural exchange activities, information was gathered through participant reflection, documentation, and observation. The results show that the program effectively combined the principles of character education with conventional performing arts. Thai students showed greater appreciation, tolerance, respect, and empathy for cultural diversity as well as a deeper comprehension of the Pendet Dance's meaning, symbolism, and cultural value. In addition, Indonesian students who served as facilitators saw improvements in their discipline, self-assurance, cultural adaptability, and communication abilities. The exercise emphasizes how traditional performing arts can be a powerful tool for international engagement and cultural diplomacy. All things considered, the program shows that cultural exchange programs can make a significant contribution to global character education, and it is advised that comparable projects be maintained and developed in subsequent international education partnerships.

Higher education institutions are experiencing accelerated transformation driven by global competition, technological advancements, and shifting learner expectations. In this context, Internal Quality Assurance Cells (IQACs) serve a strategic role in enhancing teaching quality, strengthening learning processes, and ensuring curriculum relevance through the implementation of Outcome-Based Education (OBE) and pedagogical innovation. This study employs qualitative content analysis by synthesizing theoretical frameworks and recent empirical literature to develop an integrated model of IQAC-led academic transformation. The analysis highlights three key mechanisms: curriculum intelligence systems, pedagogical innovation ecosystems, and comprehensive faculty empowerment. Findings indicate that effective IQACs function not merely as compliance bodies but as catalysts of academic excellence by aligning curriculum with industry needs, fostering continuous faculty development, and promoting technology-enhanced learning and learning analytics. Moreover, phased OBE implementation—comprising capacity building, curriculum redesign, instructional support, and continuous evaluation—significantly strengthens learning outcomes and cultivates reflective academic culture. The proposed framework underscores the interconnected role of curriculum, pedagogy, and faculty capabilities as foundations for institutional excellence. Ultimately, the study affirms that IQACs’ strategic leadership in embedding OBE, leveraging educational technologies, and empowering faculty positions them as pivotal agents in driving sustainable educational transformation.

This study examines how equity-driven curriculum design can catalyze transformative leadership within multicultural school systems. As global classrooms become increasingly diverse, traditional curricula often fail to address disparities experienced by marginalized learners. Through qualitative library research and content analysis, this study synthesizes theoretical, empirical, and policy-oriented perspectives to explore how culturally responsive curricula promote inclusive learning ecosystems. The findings highlight that equity-driven curriculum design requires not only the integration of diverse cultural perspectives but also the dismantling of hidden power structures embedded in instructional practices. Transformative leadership emerges as a crucial driver, demanding cultural competence, collaborative decision-making, and critical pedagogical reflection from school leaders. The study emphasizes that effective equity implementation depends on teacher empowerment, distributed leadership models, and coherent policy frameworks that align curriculum, assessment, and institutional values. Moreover, technology-enhanced learning, accessibility principles, and Universal Design for Learning strengthen equity outcomes in increasingly digital environments. Overall, this research underscores the interconnectedness of curriculum, leadership, and policy in fostering democratic, socially just educational systems that affirm students’ cultural identities, reduce achievement gaps, and empower communities. The study concludes that achieving equity requires systemic transformation grounded in cultural responsiveness, shared accountability, and continuous professional development.

This research focused on improving fine motor development in early childhood through creative learning activities that utilize recycled materials, designed and implemented using the ADDIE instructional model. The study involved children from class B2 at TK Prawidya Darma and was conducted systematically through five stages, namely analysis, design, development, implementation, and evaluation. Throughout the learning process, children were engaged in a variety of hands-on craft activities, including cutting, folding, gluing, and assembling recycled objects, which were intentionally structured to train hand–eye coordination, finger flexibility, and concentration. The results demonstrate that learning activities based on recycled materials offer rich and meaningful experiences for young learners. Children showed noticeable improvements in accuracy, independence, and creative expression while completing tasks. In addition, the use of recycled materials helped cultivate environmental awareness from an early age, as children learned the value of reusing everyday objects. The classroom atmosphere became more interactive and enjoyable, with students actively participating and showing enthusiasm during the activities. Overall, the findings indicate that creative recycling-based activities are effective in supporting fine motor skill development through enjoyable, practical, and environmentally friendly learning experiences. This study concludes that integrating recycled materials into creative activities can be an innovative, sustainable, and pedagogically valuable approach for early childhood education programs.

The diversity of Indonesian society, which consists of various ethnic groups, religions, cultures, and races, represents a great potential for the creation of a harmonious social life. However, the reality shows that intolerance and narrow fanaticism still often arise and even begin to influence students in the educational environment. This has the potential to create social conflicts and damage the nation's character in the future. This study aims to examine the role of education in shaping students to avoid fanatic behavior and social conflicts. The research method used is descriptive qualitative with data collection techniques through relevant literature studies. The results of the study show that schools have a strategic role in instilling values of tolerance, inclusive attitudes, and religious moderation through learning, teacher role models, and a school culture that respects diversity. In addition, synergy between schools, families, and communities is needed to create an educational environment that is conducive to the development of students' character. Through the implementation of multicultural education and the strengthening of national values on a sustainable basis, students are expected to become a generation with strong character, tolerant, and able to live peacefully alongside others in a pluralistic society.

This study examines the emerging paradigm of federated intelligence architectures as a secure, privacy-preserving, and scalable foundation for data-driven innovation across AI, IoT, and cloud ecosystems. With billions of interconnected devices generating massive heterogeneous data, traditional centralized machine-learning models face critical limitations, including privacy risks, regulatory constraints, latency, and single points of failure. Through a qualitative content-analysis approach, this paper synthesizes contemporary research on federated learning, blockchain integration, zero-trust governance, and edge intelligence to formulate a comprehensive understanding of distributed AI infrastructures. The findings highlight that federated learning enables collaborative model training without exposing raw data, significantly enhancing privacy, security, and compliance. Moreover, combining blockchain with federated learning strengthens auditability, model integrity, and trust, while zero-trust principles provide continuous verification and adaptive security enforcement across devices. Edge-AI integration further reduces latency and bandwidth consumption, enabling real-time analytics in resource-constrained IoT environments. Collectively, these elements contribute to the formation of cognitive ecosystems capable of autonomous, interoperable, and context-aware operations. The study underscores the transformative potential of federated intelligence while identifying critical gaps that inform future research trajectories.

This study conceptually examines a self-supervised multi-scale fusion framework designed to enhance accuracy and computational efficiency in medical image segmentation, a domain where data scarcity and annotation cost remain major challenges. Traditional supervised approaches are constrained by their reliance on extensive labeled datasets, limiting applicability in real-world clinical environments. Self-supervised learning (SSL) mitigates this issue by extracting supervisory signals directly from unlabeled data, enabling the model to learn rich feature representations without human annotation. Simultaneously, multi-scale fusion architectures integrate global contextual information with fine-grained local features, supporting robust segmentation across varying anatomical structures and image resolutions. Through a qualitative methodology involving library research and content analysis, this study synthesizes state-of-the-art SSL-driven segmentation techniques and highlights how adaptive multi-scale fusion mechanisms address limitations of existing convolutional and transformer-based architectures. The analysis indicates that combining SSL and multi-scale strategies leads to more generalizable, scalable, and computationally efficient segmentation pipelines suitable for diverse medical imaging modalities. The proposed framework represents a promising direction for developing next-generation diagnostic tools capable of handling sparse labels, complex textures, and real-time deployment constraints.

This study looks at how the Contextual Teaching and Learning (CTL) model is used in Christian Religious Education (CRE) and checks how it helps students develop their critical thinking abilities. In today’s education system, students are not just expected to remember Bible stories, but also to understand what they mean, think about the moral lessons, and use Christian values wisely in their everyday lives. However, past observations show that CRE classes often use a lot of lectures and memorization, which doesn’t give students much chance to think deeply or link what they learn to real-life situations. The CTL model provides a better way for learning by using seven key parts: constructivism, inquiry, questioning, learning community, modeling, reflection, and real-world assessmen. A descriptive qualitative design was employed involving CRE teachers and students selected purposively from classes that have integrated CTL in online and face-to-face settings. Data were collected through classroom observations, semi-structured interviews, and instructional documents. The analysis followed Miles and Huberman’s stages of data reduction, display, and conclusion drawing, supported by triangulation to strengthen credibility. The findings reveal that CTL enhances student engagement, promotes higher-order reasoning, and supports the internalization of Christian values in real-life contexts. Supporting factors include a positive school climate, parental involvement, and teacher readiness, whereas obstacles emerge from limited facilities, large class sizes, and insufficient teacher understanding of CTL. The study underscores the need for context-based and character-oriented instructional designs in CRE.

This study develops an architectural design framework integrating fun and playful spatial characteristics with futuristic, science-based approaches to enhance children’s engagement with science and technology learning. Traditional learning environments are often rigid and fail to stimulate curiosity, especially among children aged 7–11. Utilizing a qualitative conceptual analysis, this research synthesizes architectural theory, neuroscience learning principles, children’s spatial psychology, and science-based design methodologies. The results show that playful spatial elements—dynamic forms, color contrasts, sensory interactions, and imaginative environments—significantly improve motivation, exploration, and conceptual understanding. Futuristic design components further support scientific literacy by projecting technological possibilities and integrating interactive digital tools. The study produces a conceptual architectural model linking playfulness, scientific experience, and future-oriented visualization. It concludes that learning environments must be intentionally designed to merge emotional engagement, cognitive stimulation, and technological representation. The implications extend to the design of children’s museums, science centers, and early STEM learning facilities.

This study explores the major challenges faced by moral education in the modern era, particularly within Madrasah Ibtidaiyah, where rapid technological advancement, globalization, and shifting patterns of student interaction have significantly influenced the traditional process of moral development. The purpose of this research is to analyze teachers’ strategies in revitalizing moral education, identify the supporting and inhibiting factors in its implementation, and evaluate its impact on students’ discipline. Employing a qualitative approach with a case study design, data were collected through in-depth interviews with the school principal, Islamic ethics teachers, and students, as well as observations of daily moral habituation practices within the school environment. The findings indicate that effective revitalization of moral education occurs through an integration of teacher role modeling, consistent habituation of moral practices, and value internalization supported by interactive learning methods. These strategies contribute to improved student discipline, greater independence, and more stable behavioral patterns. However, the effectiveness of these efforts is still influenced by external factors such as excessive gadget use and limited continuity of moral guidance at home. The study provides an integrative and adaptive model of moral education that aligns with the characteristics of the digital generation and emphasizes the essential collaboration between teachers, schools, and families. The results highlight the importance of a responsive, contextually relevant approach to moral development that meets the evolving needs and challenges of 21st-century learners.

This systematic literature review examines research trends in problem-based learning (PBL) within blended learning contexts by analyzing 64 peer-reviewed articles from Scopus database published between2020-2024. Using PRISMA methodology, this study identifies key patterns including the dominance of medical education contexts (71.9%), the significant impact of COVID-19 pandemic on pedagogical innovations, and the prevalence of experimental research designs (34.4%). Findings reveal that blended learning emerged as the most researched approach (61%), followed by online/distance learning (23%) and flipped classroom models (11%). The analysis indicates increasing publication trends in 2022 and 2024, with Education Sciences and BMC Medical Education as leading publication venues. This review highlights research gaps in engineering and technology education contexts, limited exploration of long-term learning outcomes, and insufficient mixed-methods studies. These findings provide valuable insights for educators, researchers, and policymakers in designing effective learning strategies that integrate problem-based and blended learning approaches in diverse educational settings.

This Community Service (PKM) activity aims to improve artificial intelligence (AI) literacy in students majoring in Visual Communication Design (DKV) at SMK Muhammadiyah 2 Palembang through Canva-based training. The development of AI technology has brought about a significant change in the world of graphic design, where features such as Magic Design, Text-to-Image, and Background Remover are now widely used to support the creative process. However, most vocational school students are still passive users, lacking an understanding of the AI principles behind the technology. Through a Project-Based Learning (PjBL) and Active Learning approach with a demo–practice–reflection model, this activity is designed so that students are not only able to use the AI features in Canva, but also understand the work concepts, benefits, and ethics of using them. The activity was conducted over two days, involving 30 DKV class XI students. The evaluation of the results showed a significant increase in the aspects of understanding AI concepts (48%), motivation to learn design (35%), and creativity of works (32%). Students also show an increase in learning independence and digital ethical awareness. These results confirm that AI-based learning through Canva is effective in building AI literacy, creativity, and student readiness to face the technology-based creative industry. This activity is recommended to be expanded to teachers and other departments to strengthen digital literacy in the vocational education environment.  

This study aims to develop learning media in the form of an Artificial Intelligence-based maze game on geometric transformation material. The development of this learning media is motivated by the fact that mathematics learning tends to be boring, especially on geometric transformation material. This occurs because the application of monotonous learning methods, such as lecture methods, causes students to tend to be more passive and can lead to a decrease in student learning motivation. So the researcher developed a more interesting learning media in the form of an AI-based maze game to build student learning motivation. The game developed contains geometric transformation questions and presents them adaptively through AI logic. This research method uses an R&D (Research and Development) approach with the ADDIE model. The ADDIE stages are Analyze, Design, Development, Implementation, and Evaluation. The subjects of this study were ninth-grade students of SMP Yayasan Pendidikan Ghafa Medan. The results showed that this Artificial Intelligence-based maze game obtained a validity percentage of 82% with a very valid category, practicality obtained a percentage of 85.6% with a very practical category, and effectiveness obtained a percentage of 83.1% with a very effective category. The final evaluation showed that this learning tool was suitable for use, although improvements were still needed, such as adding penalties for wrong answers and enlarging the movement buttons. Overall, this AI-based maze game was deemed valid, effective, and practical as an interactive learning tool for geometric transformations.

This research examines the mentoring of nationalistic activities by students from the Al-Azhar Menganti Gresik Institute and field supervisors at MI Darunnajah Lebani Suko, Wringinanom District, Gresik Regency, in commemoration of Heroes' Day. Thru a descriptive qualitative approach, this program involves four main activities: the red and white parade, patriotic love songs, little warrior strokes (coloring and writing hero stories), and a colossal drama of patriotic fervor. These activities are creatively designed and based on direct experience to instill nationalistic values in elementary school students. The research results show a significant improvement in students in terms of courage, discipline, creativity, cooperation, and respect for national symbols such as the flag and national anthem. Collaboration between students as facilitators, teachers as pedagogical directors, and the school environment creates a conducive learning ecosystem where participatory methods are more effective than conventional approaches. This experiential learning approach not only enhances cognitive understanding of the struggles of national and local heroes but also builds emotional empathy and social skills. Positive impacts include high student enthusiasm, reduced learning boredom, and character strengthening in accordance with the Pancasila Student Profile, such as mutual cooperation and integrity. Overall, this mentoring program serves as a relevant model for non-formal character education that can be replicated in other schools to strengthen nationalism from an early age amidst the challenges of post-pandemic globalization and individualism.

This study aims to develop interactive learning media in the form of a drag-and-drop game of coordinate points on a Canva-based Cartesian plane to improve the understanding of Analytical Geometry concepts in grade VII students of MTs Yayasan Madrasah Islamiyah Sinaksak. This study applies the Research and Development (R&D) method using the ADDIE model, which consists of the stages of Analysis, Design, Development, Implementation, and Evaluation, and involves grade VII students in the odd semester of the 2024/2025 Academic Year as research subjects. Data collection instruments include expert validation sheets that assess aspects of material, language, and media design; student response questionnaires covering aspects of preferences, enthusiasm, willingness to learn, understanding, and motivation; and learning outcome tests in the form of pre-tests and post-tests. Feasibility analysis was carried out using a Likert scale, while media effectiveness was measured by comparing actual scores and ideal scores and improvements in learning outcomes were analyzed using the N-Gain formula. The results of the study indicate that the developed drag-and-drop media is categorized as feasible based on the validator's assessment, received a positive response from students, and is able to improve learning outcomes in the moderate to high category based on the N-Gain value. Thus, this media is declared effective and can be used as an innovative alternative in learning Cartesian coordinates at the MTs level.

This study examines the role of teachers as influencers in 21st-century education, particularly in developing students’ intelligence and character through the use of digital media and technology. The digital transformation requires teachers to serve not only as transmitters of knowledge but also as public figures capable of providing positive influence through digital behavior and communication. The purpose of this study is to analyze how teachers perform their roles as educational influencers and how this impacts students’ intellectual and character development. This research employs a literature study method by reviewing scientific sources such as books, articles, journals, and relevant research reports. The analysis was conducted using a descriptive-analytical approach to identify patterns, ideas, and strategies used by teachers in integrating character values and digital literacy. The findings indicate that teachers with strong digital competence and personal integrity are able to create positive influence through educational content, digital role modeling, and ethical communication. The implications highlight the importance of strengthening teachers’ digital literacy, educational policy support, and continuous professional training to create a digital learning ecosystem that is character-based, adaptive, and responsive to contemporary challenges.