Publication Search

This study aims to improve student engagement and learning outcomes in the subject of Natural and Social Sciences (Ilmu Pengetahuan Alam dan Sosial / IPAS) through the implementation of the Auditory, Intellectually, Repetition (AIR) learning model in a fourth-grade elementary school class. The initial problem indicated that student engagement in learning was still low, at 37.5%, with learning mastery reaching only 33.3% and an average class score of 68.0, which did not meet the Minimum Mastery Criteria (KKM) of 75. Therefore, improvement efforts were needed through the implementation of a more active and student-centered learning model. This study employed a Classroom Action Research (CAR) approach conducted in two cycles, where each cycle consisted of planning, action, observation, and reflection stages. The research subjects were 24 fourth-grade students. Data collection techniques included observation of student engagement, learning outcome evaluation tests, field notes, and documentation. Student engagement data were analyzed using percentages, while learning outcomes were analyzed through mean scores and the percentage of classical learning mastery. The results showed a significant improvement in each cycle. In Cycle I, student engagement increased to 62.5%, with learning mastery reaching 54.17% and an average score of 74.29, although it had not yet achieved classical completeness. In Cycle II, student engagement increased to 87.5%, with learning mastery reaching 100% and an average score of 85.42. These improvements indicate that the implementation of the AIR model was able to gradually and sustainably enhance both the learning process and outcomes. Based on these findings, it can be concluded that the Auditory, Intellectually, Repetition (AIR) learning model is effective in improving student engagement and learning outcomes in IPAS. This model can serve as an alternative learning strategy to create a more active, systematic, and student-centered learning environment.

This research is motivated by the urgency of effective Islamic Education Management (MPI) as a prerequisite for enhancing the role of Islamic Education (PAI) teachers in shaping students’ religious character amid the challenges of digitalization and institutional coordination constraints. The purpose of this study is to analyze the implementation of MPI experienced by PAI teachers, the interaction among stakeholders, and the teachers’ perceptions of its effectiveness at PKPPS Minhajurrosyidin, East Jakarta. Using a descriptive qualitative method with a case study and phenomenological approach, data were collected through triangulation (interviews, observations, and document studies) and analyzed using the Miles and Huberman interactive model. The findings indicate that MPI is an essential structural mediating variable. The optimization of MPI—particularly in terms of digital facility support and professional training—significantly enhances the effectiveness of PAI teachers as Uswah Hasanah and as facilitators of adaptive and scientific learning. However, the main obstacle lies in the lack of horizontal coordination among teachers in holistically integrating Values-Based Education. The implications of this study emphasize the need for formal managerial policies to ensure collective responsibility in character formation.

This study aims to develop a storytelling-based learning media in the form of a YouTube video to support students’ understanding of narrative text based on their learning needs. This study employed a Research and Development (R&D) design using the ADDIE model, focusing on the stages of analysis and development. The data were collected through a questionnaire as part of the needs analysis to identify students’ responses toward listening and repetition activities, as well as the difficulties they face in understanding narrative texts. The results show that students generally respond positively to listening and repetition. However, they still experience difficulties in understanding the storyline and vocabulary. The findings also indicate that students need learning support in the form of audio, visual elements, and vocabulary assistance. Based on these findings, a learning product was developed in the form of a YouTube-based digital storytelling video entitled “Learn Narrative Text Through Fun Storytelling (Listen, Repeat & Retell)”. The video integrates storytelling, repetition, vocabulary explanation, and speaking practice to support students’ comprehension. Due to time limitations, this study was limited to the development of a prototype and has not yet included expert validation. Therefore, future research is recommended to conduct validation to evaluate the feasibility and effectiveness of the developed product.

The retention of customers in the retail banking sector is a critical economic imperative; however, predictive modeling is frequently hindered by severe class imbalance and the “Black Box” nature of complex algorithms. This study proposes a Heterogeneous Stacking Ensemble framework integrating XGBoost, CatBoost, and Random Forest base learners with a Logistic Regression meta-learner to forecast customer attrition. To overcome the pervasive “Majority Class Bias,” we introduce a “Dual-Imbalance Defense” that synergizes the Synthetic Minority Over-sampling Technique (SMOTE) with algorithmic cost-sensitive penalization. Furthermore, moving beyond standard accuracy metrics, the framework mathematically derives a dynamic classification threshold to guarantee a strict 0.90 recall rate, actively optimizing the capture of at-risk capital. Model opacity is addressed through the integration of a SHapley Additive exPlanations (SHAP) TreeExplainer. This cooperative game theory approach provides localized, patient-level “Reason Codes” for regulatory compliance and reveals global systemic vulnerabilities, including non-linear drivers such as the “Product Paradox.” Achieving a 0.90 recall rate and an AUC of 0.8654, this framework provides a statistically robust and operationally transparent tool for targeted customer retention.

This study aims to develop and implement an artificial intelligence-based reading learning application using Deep Learning technology to enhance the literacy skills of eighth-grade junior high school students. The research employed the Kemmis & McTaggart Classroom Action Research model combined with a mixed-methods approach. Data collection involved pretests and posttests, complemented by observations, interviews, and questionnaires. The findings revealed that the use of this application significantly improved students' reading comprehension, question-answering skills, and overall engagement in the learning process. Key features of the application, such as adaptive learning technology, allowed for real-time adjustments to the difficulty level of the material, which catered to each student’s individual learning pace. Additionally, the provision of instant feedback enhanced the learning experience by helping students understand their progress and areas for improvement. These results suggest that the application is an effective tool in fostering literacy development and aligns with the goals of the Independent Curriculum. Consequently, this Deep Learning-based application offers a promising innovation for improving student literacy skills in the digital age. 

Object detection plays a crucial role in intelligent transportation systems, particularly for outdoor traffic monitoring applications that require accurate and real-time performance under limited computational resources. Recent developments in YOLO-based architectures have introduced multiple model variants; however, their practical performance under constrained training conditions remains insufficiently explored. This study presents a comparative evaluation of YOLOv5, YOLOv7, and YOLOv8 for outdoor traffic object detection using a real-world dataset and identical experimental settings. The main objective of this research is to analyze the robustness and detection quality of different YOLO variants when trained with a limited number of epochs, reflecting practical deployment scenarios. All models were trained and evaluated using the same dataset, preprocessing pipeline, and hardware configuration to ensure a fair comparison. Performance evaluation was conducted using multiple metrics, including precision, recall, mAP@50, Precision–Recall curves, area under the curve (AUC), and peak F1-score. Experimental results indicate that YOLOv5 outperformed YOLOv7 and YOLOv8 in terms of overall detection stability and robustness. The merged Precision–Recall analysis shows that YOLOv5 achieved a higher effective AUC and superior mAP@50, reflecting better global detection performance. In addition, YOLOv5 exhibited a higher peak F1-score, indicating a more balanced trade-off between precision and recall. In contrast, YOLOv7 and YOLOv8 showed performance degradation under limited training conditions despite their more advanced architectures. These findings suggest that YOLOv5 remains a reliable and efficient solution for outdoor traffic object detection, particularly in resource-constrained environments. The study highlights the importance of comprehensive evaluation metrics and practical experimental settings when selecting object detection models for real-world applications.

Object detection plays a crucial role in intelligent transportation systems, particularly for outdoor traffic monitoring applications that require accurate and real-time performance under limited computational resources. Recent developments in YOLO-based architectures have introduced multiple model variants; however, their practical performance under constrained training conditions remains insufficiently explored. This study presents a comparative evaluation of YOLOv5, YOLOv7, and YOLOv8 for outdoor traffic object detection using a real-world dataset and identical experimental settings. The main objective of this research is to analyze the robustness and detection quality of different YOLO variants when trained with a limited number of epochs, reflecting practical deployment scenarios. All models were trained and evaluated using the same dataset, preprocessing pipeline, and hardware configuration to ensure a fair comparison. Performance evaluation was conducted using multiple metrics, including precision, recall, mAP@50, Precision–Recall curves, area under the curve (AUC), and peak F1-score. Experimental results indicate that YOLOv5 outperformed YOLOv7 and YOLOv8 in terms of overall detection stability and robustness. The merged Precision–Recall analysis shows that YOLOv5 achieved a higher effective AUC and superior mAP@50, reflecting better global detection performance. In addition, YOLOv5 exhibited a higher peak F1-score, indicating a more balanced trade-off between precision and recall. In contrast, YOLOv7 and YOLOv8 showed performance degradation under limited training conditions despite their more advanced architectures. These findings suggest that YOLOv5 remains a reliable and efficient solution for outdoor traffic object detection, particularly in resource-constrained environments. The study highlights the importance of comprehensive evaluation metrics and practical experimental settings when selecting object detection models for real-world applications.

 Low student thesis completion productivity is a challenge in higher education, particularly at UIN Palangka Raya. Thesis writing requires self-regulation skills and time discipline to enable students to complete their final assignments effectively and on time. This study aims to analyze the role of self-regulated learning and time discipline in improving the thesis completion productivity of final-year students. The study used a qualitative approach with a case study design of five final-year students who were in the process of completing or had completed their theses. Data collection techniques included in-depth interviews, limited observation, and documentation. The data were then analyzed using the Miles and Huberman interactive model through the stages of data reduction, data presentation, and conclusion drawing. The results indicate that self-regulated learning plays a role in helping students plan goals, control motivation and emotions, and conduct consistent self-evaluation. Time discipline has been proven effective in reducing procrastination through the implementation of daily schedules, prioritization, and distraction management. Therefore, the integration of self-regulated learning and time discipline is an important strategy in increasing the thesis completion productivity of students and supporting sustainable academic success.

The digital era demands a fundamental transformation in the role of early childhood educators, shifting from passive technology consumers to active architects of digital literacy. However, the dominance of technocentric views often acts as a substantial psychological and pedagogical barrier for teachers in regional areas. This collaborative community service project aims to reconstruct the paradigm of 50 kindergarten teachers in Bulukumba Regency by integrating "unplugged coding" logic and deep learning into play-based learning. Utilizing a Product-Based Intensive Training method with a "Logic over Laptop" strategy, the program focused on deconstructing technology-related stigmas and reconstructing teachers' ability to transform abstract concepts into safe, concrete media for children. Data analysis revealed a significant shift in teacher paradigms; while the majority were initially in the "less successful" category, 100% of participants reached positive categories (successful and very successful) post-intervention. Statistically, the program's effectiveness was evidenced by a dramatic increase in mean scores from 18.04 to 31.24 (p < 0.05) and an N-Gain score of 0.778, classified as highly effective. Furthermore, the partner satisfaction index reached 4.82 (very satisfied), confirming that the tri-campus collaboration model (STAI Al-Gazali, UNM, and Unismuh) is highly relevant to the implementation of the Merdeka Belajar curriculum. This project concludes that strengthening digital literacy through non-digital algorithmic reasoning effectively dismantles technical barriers for teachers while ensuring the safety of child development in the digital age.

This study aims to identify the learning needs of deaf students in internship programs, examine the challenges they face, develop appropriate solutions, and design as well as evaluate a visual module-based learning model to improve their work skills. The study used a Research and Development (R&D) approach with a 4D model: Define, Design, Develop, and Disseminate. The participants included deaf students from special needs high schools (SMALB) involved in vocational internships, mentor teachers, and industry supervisors. Data were collected through observation, interviews, questionnaires, documentation, and focus group discussions, and analyzed using qualitative techniques supported by descriptive analysis. The findings indicate that deaf students require visual, structured, and easily understandable work instructions supported by symbols, color codes, and guidance materials. Major challenges include limited verbal communication, difficulty understanding instructions, and risks of procedural errors. To address these issues, a systematic and communicative visual module-based learning model was developed, incorporating collaborative support from schools and industry. The resulting model integrates planning, implementation, mentoring, and evaluation stages, and has proven feasible and effective in enhancing students’ independence, technical competence, and overall work readiness.

Attention mechanisms have been widely incorporated into recurrent neural network architectures for financial time series forecasting, with most prior work reporting improvements in price-level error metrics. This study revisits that claim through a controlled empirical comparison of four deep learning architectures on nearly two decades of Telkom Indonesia (TLKM) closing price data from the Indonesia Stock Exchange (IDX). The models evaluated are a three-layer Gated Recurrent Unit (GRU) baseline, a comparable Long Short-Term Memory (LSTM) network, a Bahdanau end-attention GRU (Attn-GRU-V2), and a multi-head self-attention GRU hybrid (Attn-GRU-V3). Each architecture is trained over 30 independent runs with distinct random seeds, and performance is reported as 95% confidence intervals derived from the t-distribution. Statistical comparisons employ the Wilcoxon signed-rank test, a nonparametric paired test appropriate given the confirmed non-normality of residuals. The main finding is a consistent trade-off: the plain GRU achieves the lowest RMSE (94.02 ± 1.22 IDR) across all 30 runs, while Attn-GRU-V2 achieves the highest directional accuracy (45.91 ± 0.09%), surpassing GRU in every independent run. Bahdanau attention weights are nearly uniform across the 30-day lookback window (coefficient of variation: 3.21%), indicating that the mechanism cannot identify selectively informative timesteps in this univariate price series. This finding is consistent with the weak-form Efficient Market Hypothesis for the Indonesian market. An ablation study reveals that a 20-day lookback window maximizes directional accuracy (47.72 ± 0.21%) for the Attn-GRU-V2 model. These results suggest that Bahdanau end-attention consistently and significantly improves directional accuracy relative to a plain GRU baseline, providing an architecturally attributable advantage for direction-based applications, even when absolute price-level error is not reduced. The directional accuracy values remaining below 50% across all models are consistent with a weak-form efficiency characterization of the Indonesian market.

The increasing sophistication of malware has rendered traditional signature-based detection methods insufficient, necessitating behavior-driven and adaptive analytical frameworks. This study presents a sequential deep learning framework that models system-level API call sequences as structured linguistic representations for behavioral malware detection. Unlike conventional comparative studies, this work systematically evaluates recurrent and attention-based architectures under controlled experimental conditions, with a particular focus on generalization performance and overfitting mitigation. Two neural architectures, a Long Short-Term Memory (LSTM) network and a Transformer-based attention model, are trained on publicly available API call sequence data for binary classification of malicious and benign executables. Beyond standard accuracy metrics, the study further examines model stability, convergence behavior, and the impact of long-range dependency modeling on detection robustness. Experimental results demonstrate that the Transformer architecture achieves superior performance, attaining 95.54% classification accuracy and consistent improvements in precision, recall, and F1-score, indicating a stronger ability to capture complex behavioral dependencies. These findings highlight the effectiveness of attention mechanisms in behavioral malware modeling and provide empirical evidence that NLP-inspired architectures offer a robust and scalable approach for real-world cybersecurity applications.

This study aims to evaluate the impact of leadership education and training on organizational service effectiveness within the Tambrauw Regency Government, Papua. Leadership training is an important instrument in developing the competencies of Civil Servants to enhance leadership capacity, managerial skills, and public service quality. However, its implementation needs evaluation to determine its impact on organizational performance and service effectiveness. This research uses a qualitative approach with descriptive analysis. The analytical framework applied is the Kirkpatrick training evaluation model, covering reaction, learning, behavior, and results. Data collection was conducted through documentation studies, policy analysis, and literature review on civil servant development and public services. The results indicate that leadership training has a positive impact on improving civil servant competencies and service effectiveness. At the reaction level, participants showed high satisfaction with training materials and methods. At the learning level, there was improvement in leadership knowledge and skills. At the behavior level, participants demonstrated more professional work attitudes, better coordination, and improved decision-making. At the results level, training contributed to improved service quality, efficiency, and innovation. However, challenges remain, including limited resources, hierarchical bureaucratic culture, and weak policy support. Therefore, strong local government commitment is needed to enhance training quality and organizational support. This study contributes to public administration research and offers practical insights for policy development.

The rapid advancement of digital technology in higher education has transformed learning and assessment systems, particularly through the adoption of Learning Management Systems (LMS) and Computer-Based Tests (CBT). However, many university students still face challenges in adapting to digital examination environments due to limited technical experience and digital literacy. This Community Service Activity (PKM) aims to enhance students’ academic readiness and digital competence through the implementation of a Moodle-based LMS at Universitas Murni Teguh. The activity was conducted in the form of an interactive workshop, including system introduction, simulation of CBT, guided practice, and discussion sessions involving university students and lecturers. The results showed a significant improvement in students’ understanding of LMS features, confidence in using CBT systems, and ability to manage digital-based examinations effectively. Active participation and engagement during the sessions reflected high enthusiasm and adaptability toward digital learning transformation. This activity demonstrates that the implementation of LMS Moodle is effective in improving academic preparedness and digital literacy among students, and can serve as a sustainable model for strengthening technology-based learning in higher education institutions.

The rapid advancement of digital technology in higher education has transformed learning and assessment systems, particularly through the adoption of Learning Management Systems (LMS) and Computer-Based Tests (CBT). However, many university students still face challenges in adapting to digital examination environments due to limited technical experience and digital literacy. This Community Service Activity (PKM) aims to enhance students’ academic readiness and digital competence through the implementation of a Moodle-based LMS at Universitas Murni Teguh. The activity was conducted in the form of an interactive workshop, including system introduction, simulation of CBT, guided practice, and discussion sessions involving university students and lecturers. The results showed a significant improvement in students’ understanding of LMS features, confidence in using CBT systems, and ability to manage digital-based examinations effectively. Active participation and engagement during the sessions reflected high enthusiasm and adaptability toward digital learning transformation. This activity demonstrates that the implementation of LMS Moodle is effective in improving academic preparedness and digital literacy among students, and can serve as a sustainable model for strengthening technology-based learning in higher education institutions.

This study aims to develop Smartboard Book teaching materials based on Augmented Reality that are valid, practical, and effective for improving students' learning outcomes in the IPAS (Science) subject for grade IV at SDIT Al-Farabi. This study uses the Research and Development (R&D) method with the ADDIE (Analysis, Design, Development, Implementation, Evaluation) development model. The subjects of this study were grade IV students at SDIT Al-Farabi. Data collection techniques were conducted through expert validation questionnaires, response questionnaires, and learning outcome tests. The results indicate that the Smartboard Book teaching material was declared very valid, with material and test instrument validation percentages of 88% and media validation of 95%. The teaching material was also declared practical based on teacher and student responses. Furthermore, this media proved to be effective in improving students' learning outcomes with an N-gain score of 0.86, which falls into the high category. Based on these findings, it is recommended that teachers utilize Smartboard Book teaching materials based on Augmented Reality as an alternative medium in learning activities to create a more interactive and engaging learning environment.

This study aims to descriptively analyze the ability of elementary students to imitate rhythmic gymnastics movements through the Just Dance Now application in Physical Education learning at SDN Bulay 1 Pamekasan. The background of this research stems from the low motivation and movement accuracy of students in conventional rhythmic gymnastics learning. Using a qualitative descriptive approach, this study involved 24 fifth-grade students as subjects. Data collection techniques included participatory observation, in-depth interviews, and documentation. Data were analyzed using the Miles and Huberman interactive model consisting of data reduction, data presentation, and conclusion drawing. The findings indicate that students' ability to imitate rhythmic gymnastics movements through Just Dance Now falls into the sufficient category, with students demonstrating good proficiency in basic movement patterns but experiencing challenges in complex body coordination and musical rhythm synchronization. The application successfully enhanced student enthusiasm and engagement compared to traditional methods. However, technical constraints such as internet connectivity and limited space were identified as implementation barriers. This study implies that digital game-based media like Just Dance Now can serve as an effective alternative visual aid in elementary physical education, particularly for rhythmic gymnastics instruction. Educators are encouraged to integrate technology creatively while addressing infrastructural limitations to optimize learning outcomes.

Many senior high school students face significant difficulties in generating ideas, organizing thoughts, and selecting appropriate vocabulary when writing descriptive texts. Traditional teaching methods often limit creativity, making the integration of visual media crucial. Therefore, this study aims to develop creative poster-based learning module for teaching descriptive writing. A Research and Development (R&D) design was employed, adapting the ADDIE (Analysis, Design, Development, Implementation, Evaluation) model. However, due to time constraints, this research is strictly limited to the Analysis (Needs Analysis) and the initial Development (Prototype Development) stages. The participants for the needs analysis consisted of 22 senior high school students in Indonesia, selected through purposive sampling. Data were collected using an online needs analysis questionnaire. The findings indicate a strong student preference for visual support, which informed the design and creation of a poster-based learning module prototype intended to help students visualize concepts. The implications suggest that while the prototype aligns with student needs, further research is required to validate and test the product.

The rapid growth of digital marketing has increased the demand for a competent workforce, while many vocational school teachers still face limitations in both theoretical understanding and practical implementation of digital marketing practices. This community service program aimed to strengthen the digital marketing competencies of 15 teachers at SMKN 1 Purwokerto through workshops facilitated by certified lecturers and industry practitioners. The program applied a participatory and collaborative approach consisting of needs analysis, workshop sessions, curriculum realignment, industry-based module development, and continuous mentoring over six months. The findings showed significant improvements in teachers’ competencies, with an 82.6% increase in digital marketing comprehension scores, rising from 45.3 to 82.6. In addition, all participating teachers successfully mastered innovative lesson planning and the implementation of digital marketing tools in classroom learning. The program also produced eight validated industry-based learning modules, exceeding the initial target. Furthermore, 87% of teachers who were interested in professional certification successfully obtained certification. The results indicate that collaboration between academics and industry practitioners effectively enhances vocational teachers’ competencies and provides a replicable model for improving the quality and industry relevance of vocational education in the digital era.

This study examines the integration of Islamic education values in supporting the achievement of the Sustainable Development Goals (SDGs) 2030. The background of this research is rooted in the increasing global demand for education systems that not only emphasize cognitive competence but also foster moral integrity, social responsibility, and environmental awareness. Islamic education, with its foundational values such as tawhid (divine consciousness), adl (justice), amanah (responsibility), and mizan (balance), offers a holistic framework that aligns with the principles of sustainable development. The objective of this research is to analyze how Islamic educational values can be conceptually and operationally integrated into educational practices to contribute to the SDGs agenda. This study employs a qualitative approach using library research as the primary method, collecting and analyzing relevant academic literature published in recent years. The findings indicate that the integration of Islamic values through contextual learning, reflective discussions, character-building activities, and social engagement initiatives can enhance students’ awareness of global issues, empathy, inclusivity, and ecological responsibility. Although challenges remain in curriculum standardization and character assessment mechanisms, the study concludes that Islamic education has strong potential to function as a transformative educational model. The implications of this research suggest the need for curriculum development, teacher capacity building, and institutional commitment to ensure that Islamic education meaningfully contributes to sustainable global development.