Publication Search

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 problem in this study is to determine what is the effect of the number tube game on the symbolic thinking ability of 4-5 year olds. The aim is to reveal whether there is a number tube game on the symbolic thinking ability of 4-5 year old children at Menara Laut State Kindergarten, Leato Utara Village, Dumbo Raya District, Gorontalo City. The type of research used is a quantitative experiment. The subject of this study is how to determine the effect of the number tube game. The method used in this study is the observation and documentation method. Based on the results of the observation, it was found that the symbolic thinking ability in recognizing numbers in children has not developed optimally. Although teachers have tried to introduce numbers, most children still have difficulty in naming, ordering, and writing numbers. Children also need teacher assistance to understand numbers 1-10 and distinguish between certain numbers such as 6 and 9. Therefore, a more effective and fun learning strategy is needed to help children understand the concept of numbers symbolically. In this case, it can be concluded that improvements through the application of the number tube game. This game was chosen because it is simple, easy to use, and in accordance with the characteristics of early childhood. Through a play-while-learning approach, children are expected to be able to recognize numbers more easily.

Penelitian ini mengkaji persepsi siswa terhadap penggunaan e-learning dalam pembelajaran Pendidikan Kewarganegaraan di SMAN 4 Pekanbaru. Dengan pendekatan kualitatif melalui wawancara terhadap satu guru dan kuesioner terbuka kepada sepuluh siswa kelas XI, penelitian ini menelusuri bagaimana pembelajaran digital memengaruhi pemahaman siswa terhadap nilai-nilai kewarganegaraan. Hasil penelitian menunjukkan bahwa siswa umumnya memiliki persepsi positif terhadap e-learning karena fleksibilitas dan kemudahan akses, meskipun keterbatasan interaksi langsung menjadi kendala. Hambatan lainnya termasuk koneksi internet yang tidak stabil dan motivasi belajar yang menurun. Namun, strategi pembelajaran kontekstual seperti penugasan presentasi budaya lokal terbukti mampu memperkuat nilai kewarganegaraan dan keterlibatan siswa. Penelitian ini menegaskan pentingnya komunikasi pembelajaran yang efektif dan konten digital yang berpusat pada siswa dalam pembelajaran PKn. E-learning, bila dirancang dengan baik, dapat berperan besar dalam pembentukan karakter dan kewarganegaraan digital.

This research aims to develop a web-based e-learning application at Sekolah Dasar Masehi Tanggaba by applying the Waterfall software development model. The background of this study stems from the need for interactive, flexible, and easily accessible learning media in the digital era, considering that conventional face-to-face methods are still predominantly used in schools. The Waterfall model was chosen because it provides a systematic workflow through the stages of requirements analysis, system design, implementation, testing, and maintenance. Data collection was carried out through observations, teacher interviews, and documentation studies to identify the actual needs of users.The results of this study produced an e-learning application with key features, including learning material management, interactive quizzes, discussion forums, and automated assessment. Teachers can upload structured materials and assignments, while students can access learning content, complete exercises, and interact through forums more flexibly. System testing was conducted using the black box method to ensure that each function runs according to user requirements and specifications. Based on the testing results, the application was proven to function properly, be practical to use, and effective in supporting the learning process. Teachers found it helpful in managing materials and assignments, while students experienced more independent and interactive learning that was not limited to the classroom. Therefore, the developed e-learning application can be considered valid, practical, and effective as a supporting learning medium in elementary schools and has the potential to improve learning quality in the digital era.

The purpose of this education is to increase public awareness, especially among teenagers, of the impacts of early marriage through educational seminars and Service-Learning-based outreach in Naga Kesiangan Village. Early marriage remains a critical issue in Indonesia, often associated with health risks, psychological stress, and limited educational and social opportunities for young people. The Service-Learning approach was chosen because it not only provides knowledge transfer but also actively engages participants in reflection and discussion, ensuring that the learning process is more contextual and sustainable. The method used was Service-Learning, involving 30 teenagers aged 15–19 years, 15 parents, and 5 community leaders who were purposively selected for their direct relevance to the issue. Activities included seminars, small group discussions, interactive question-and-answer sessions, and collective reflection. The results of the activity showed a significant increase in participants’ understanding, from 42% (pre-test) to 81% (post-test). Participants gained a deeper understanding of the health, psychological, social, and educational risks of early marriage and were encouraged to delay marriage by choosing positive alternatives through education and skill development. This program also strengthened the role of families and community leaders as preventive agents by fostering collective awareness and support. Thus, educational seminars integrated with Service-Learning prove to be effective and replicable strategies to prevent early marriage practices, while also empowering communities to build healthier, more resilient generations.

E-learning has become a vital pillar in higher education, especially since the COVID-19 pandemic forced a sudden shift towards digital learning methods. This article examines the effectiveness of e-learning platforms through a comparative literature review (2020–2025), focusing on Moodle, Google Classroom, Microsoft Teams, Zoom, and CIDOS 4.0 as the local platform. The study evaluates aspects of student engagement, accessibility, assessment, and technological innovation. Findings reveal that although e-learning enhances flexibility, challenges remain in terms of the digital divide, system usability, and reliability of assessment. This review highlights that Moodle provides strong interactive and assessment features, while Google Classroom offers simplicity and accessibility but lacks advanced analytics. Microsoft Teams supports collaboration and integration with Office 365, yet requires stable internet. Zoom is highly effective for synchronous interaction but less suitable for task management. CIDOS 4.0 addresses Malaysia’s TVET context but continues to face issues with server stability and user interface design. Beyond platform-specific strengths and weaknesses, the analysis emphasizes that student engagement, fair access, and reliable evaluation methods are central to e-learning effectiveness. Hybrid approaches that combine face-to-face and online learning are identified as the most effective strategy within the Malaysian higher education context. The study concludes that institutions must not only choose appropriate platforms but also enhance digital literacy, invest in infrastructure, and explore innovations such as AI and gamification. These measures can maximize e-learning’s potential to improve teaching, learning quality, and the broader digital transformation of higher education.

The development of e-learning platforms has transformed the landscape of education by providing flexible and accessible learning methods. However, security threats targeting web-based applications pose significant risks to the confidentiality, integrity, and availability of academic data. This research aims to analyze the security vulnerabilities of the ILMU2 e-learning website at UPN “Veteran” Jawa Timur using the Web Application Security Project Zed Attack Proxy (OWASP ZAP) methodology. The study applies a black-box testing approach to simulate attacks and identify potential weaknesses in the website's security mechanisms. Expected findings include the identification of vulnerabilities such as cross-site scripting (XSS), SQL injection, and security misconfigurations. The results are intended to provide insights for improving the security posture of the ILMU2 platform and contribute to the broader discourse on securing educational technology in higher education institutions.

The purpose of this research is to see how Digital Literacy: Understanding Norms and Ethics in the Online Field. In general, this study aims to assess the extent of understanding and practice of digital ethics in various groups. Measuring interventions (workshops/socialization) in improving ethical behavior online. Identify gaps in areas prone to norm violations, such as privacy breaches, unauthorized uploads, and the spread of hoaxes. Develop a model or recommendation for literacy programs that is grounded in social norms and values. Research methods The form of this research is descriptive, analytical, and critical. Therefore, the author can comprehensively explain how Digital Literacy: Understanding Norms and Ethics in the Online Field. In this study, the author optimally used two data sources related to this research, namely. Primary data sources and secondary data sources. The result of this study is that digital literacy and ethics are effective in reducing hoaxes and hate speech. Strengthening methods include e-learning, digital media, and school literacy movements. The challenges are the limited skills of students, the lack of parental supervision, and the need for teachers who need to continue to develop.

Vocational high schools (SMK) play an important role in preparing students for the world of work. However, student learning outcomes in the subject of Electrical Lighting Installation at SMK Negeri 5 Kota Padang are still relatively low, with an average pretest score of 57.34. This study aims to test the effectiveness of the Student Teams Achievement Division (STAD) cooperative learning model in improving student learning outcomes. The research method used was classroom action research with 32 students from class XI TITL as subjects. Data were collected through pre-tests, post-tests, and observations of student activities during learning. The results of the study indicate that the implementation of STAD successfully improved student learning outcomes significantly, with an average post-test score of 82.19 and an N-Gain Score of 0.58 (moderate category). Academic achievement also increased from 15.6% to 75%. Further analysis revealed that collaborative interactions within groups, the reward system, and active student participation were key factors contributing to the success of this method. However, challenges such as time management and low student engagement require attention. It is concluded that the STAD model is effective in improving students' learning outcomes in the subject of Electrical Lighting Installation, both in terms of cognitive aspects and social skills. Therefore, it is recommended that teachers implement STAD with proper syntax and consider developing supporting media such as digital simulations to optimize the learning process.

The development of digital technology has changed the landscape of educational media, especially in the field of engineering, where abstract concepts often pose a challenge. In electrical engineering education, students often have difficulty understanding concepts such as current flow and circuit behavior due to limited class time and a lack of visualization tools. This study aims to evaluate the effectiveness of Augmented Reality (AR)-based learning media in enhancing conceptual understanding, learning outcomes, and student engagement in the Electrical Circuits course. A quasi-experimental design with pre-test and post-test control groups was applied, involving two groups: one group used AR-based media, while the control group used traditional instruction. Both groups underwent pre-tests and post-tests with 30 multiple-choice questions. Data were analyzed using descriptive statistics and N-Gain scores. Results showed that the experimental group experienced a significant improvement, with the average post-test score increasing from 62 to 81 and an N-Gain score of 0.48, while the control group only improved from 61 to 71 with an N-Gain score of 0.25. Statistical tests confirmed significant differences between the two groups. These findings indicate that AR is effective in enhancing understanding of complex concepts and can be integrated into technical education curricula to meet modern learning demands.

Learners are continually faced with choosing appropriate courses or making career choices due to increased educational opportunities. The emergence of machine learning-based course and career recommender systems has the potential to address this issue, offering personalized course recommendations tailored to individual learning pathways, preferences, and learning history. The optimization and feature engineering techniques and practical deployment environments have not been collectively examined in the previous research, despite the significant advancements in this area of research. Furthermore, previous research has rarely synthesized how these technical components help students choose appropriate courses and careers. This systematic review was carried out to investigate the current state of machine learning-based course and career recommender systems, focusing on key elements, such as primary data sources, feature engineering methods, algorithms, optimization techniques, evaluation metrics, and the environments where the existing course recommendation models are deployed. The PRISMA method for conducting a systematic review was used to choose studies that met the requirements for inclusion and exclusion. The study findings show significant reliance on interpretable and traditional machine learning algorithms, such as K-Nearest Neighbor and Random Forest, to develop recommender models. Feature engineering remains basic, as most studies rely on normalization, while optimization processes are often underreported. Also, evaluation metrics varied widely, impeding comparability, while most of the recommender models are deployed in an e-learning environment, leaving the traditional learning environment underrepresented. Furthermore, the study findings identified issues including data sparsity and diversity, data security and privacy, and changes in learner preferences that may have an impact on the performance of recommender systems while recommending further studies to make use of standardized optimization methods, and automated domain-informed feature engineering frameworks, benchmark and annotated datasets in developing models the gives priority to learners’ success and educational relevance.

This article aims to understand the concept of energy sources from an early age in order to increase their awareness of energy use and conservation. However, science learning in elementary schools (SD) often faces problems in delivering complex and abstract materials. This study investigates various innovations in science learning, especially those related to energy sources. In this study, we review the literature on various creative methods for learning, including the use of game media, blended learning, and interactive game tools. The results show that the implementation of these innovations can improve students' desire to learn and their learning outcomes. Therefore, science learning at the elementary level can be assisted by the integration of technology and appropriate learning approaches.

This abstract discusses learning innovations in Islamic Religious Education (PAI) based on information technology as part of the digital transformation in Islamic education. This research aims to identify and analyze the application of information technology in the PAI learning process to improve the effectiveness, interactivity, and accessibility of Islamic education. A specific fact that is the focus is the rapid development of digital technology that influences traditional learning methods, as well as the urgent need to integrate technology into the PAI curriculum to suit the demands of the times. The research method uses a qualitative approach with case studies in several Islamic educational institutions that have implemented digital technology in PAI learning. The results of the research show that the use of information technology, such as digital learning applications, e-learning platforms and interactive multimedia, can increase student learning motivation, broaden the reach of education and support more personal and contextual learning. This research provides strategic recommendations for curriculum developers and educators to optimize the use of technology in PAI learning, while also emphasizing the importance of digital transformation as an effort to modernize sustainable Islamic education. Thus, PAI learning innovations based on information technology are key in facing the challenges and opportunities of the digital era.    

Informatics systems have an important role in improving the effectiveness of the learning process in schools. This study aims to analyze the effect of the application of informatics systems on the quality of learning. The research method used is an experimental study with a quantitative approach. Data were collected through observation, interviews, and questionnaires given to students and teachers. The results of the study indicate that the use of informatics systems, such as e-learning and technology-based academic management, can improve the efficiency of school administration and facilitate student access to learning materials. These findings indicate the importance of technology integration in the world of education.  

Quipper School E-Learning is a learning media with an E-Learning system based on open source which is a link between students and teachers in the division of tasks online. Learning using the E-Learning Quipper School media can be done anywhere, providing teaching materials according to the curriculum, students can see the value of learning otcomes directly, students and teachers can discuss through this media so that students become more active. The purpose of this study was to determine the effect of using Quipper School e-learning media on the learning outcomes of class X High School of ITCI Penajam Paser Utara on the nomenclature of chemical compounds. The type of research used is quasi-experimental research with a Post Test Only Control Group Design. The population in this study were all class X students of High School of ITCI Penajam Paser Utara for the 2023/2024 academic year. Class XI-2 as an experimental class that uses Quipper School e-learning media and Class XI-1 as a control class that does not use Quipper School e-learning media. Data collection uses tests in the form of post-tests and daily tests as well as non-tests, namely student response questionnaires, observation sheets of teacher and student activities. The results of data analysis show that the value of learning outcomes in the experimental class is 86, higher than the control class which is 81, so it can be concluded that the use of Quipper School e-learning has an effect on students' cognitive learning outcomes in thermochemical material.

Courseware is an effective and important teaching and learning model, and it is at the core of the process of creating e-learning. The objective of this paper is to show effect of quality courseware on effective e-learning. A quantitative research approach by questionnaire has been utilized. A total 90 participants have been surveyed with the use of the questionnaire. This study shows that quality courseware was the most significant positive factor and more import e learning success.

This study presents a Quantum Machine Learning (QML) architecture for perfectly classifying the Iris flower dataset. The research addresses improving classification accuracy using quantum models in machine-learning tasks. The objective is to demonstrate the effectiveness of QML approaches, specifically the Variational Quantum Circuit (VQC), Quantum Neural Network (QNN), and Quantum Support Vector Machine (QSVM), in achieving high performance on the Iris dataset. The proposed methods result in perfect classification, with all models attaining accuracy, precision, recall, and an F1-score of 1.00. The main finding is that the QML architecture successfully achieves flawless classification, contributing significantly to the field. These results underscore the potential of QML in solving complex classification problems and highlight its promise for future applications across various domains. The study concludes that QML techniques can offer transformative solutions in machine learning tasks, particularly those leveraging VQC, QNN, and QSVM.

The 21st century has witnessed a rapid digital transformation in global education, reshaping traditional learning paradigms and creating new opportunities and challenges. This study explores the impact of digitalization on educational accessibility, teaching methodologies, and student engagement. The research employs a qualitative approach, analyzing current trends, technological advancements, and policy implementations in various educational systems. The findings highlight the benefits of digital transformation, such as increased accessibility, personalized learning experiences, and the integration of artificial intelligence in education. However, challenges such as the digital divide, cybersecurity concerns, and the need for teacher training persist. The study underscores the importance of strategic policies and investment in digital infrastructure to ensure equitable and effective educational transformation worldwide.

E-Learning becomes a platform which supports technology-based learning on many aspects. The usage of E-Learning becomes the main choice especially in education world. SMK TI Bali Global Jimbaran, which is a middle level vocational school, utilizes this platform in its learning process. In its implementation process, there needs to be a precise and thorough supervision to ensure that the system runs according to business goals. This mapping is conducted to know the process used for measurements. This mapping is done by framework with COBIT 5 by applying PAM (Process Assessment Model), which is adjusted with the condition on research place. The activity measurement are derived from several domains on COBIT 5 with mapping process based on the business goals and institution goals. The domains used are EDM03, AP003, BAI07, DSS05 and MEA01. The results of governance measurement process on current capability level is on level 4 (predicable process) and the expected capability level is on level 5 (Optimizing Process) which has gap value of 2 level.

Mitigasi bencana merupakan upaya mengurangi risiko dan dampak buruk bencana melalui pembangunan fisik dan peningkatan kesiapsiagaan masyarakat. Tujuan utama program PKM ini adalah untuk mempersiapkan siswa-siswa Sekolah Dasar dari 7 sekolah di pesisir pantai Kota Bandar Lampung agar dapat menyelamatkan diri saat terjadi bencana alam dengan cara memberikan pengetahuan tentang mitigasi bencana dan menyediakan jalur evakuasi yang jelas. Lokasi percontohan untuk sosialisasi dan pelatihan adalah di SDN 1 Sukamaju Teluk Betung Timur. Metode kegiatan PKM ini yaitu Service-Learning yang merupakan pendekatan pembelajaran yang mengintegrasikan kegiatan akademik dengan pengabdian masyarakat. Hasil kegiatan berjalan lancar dan kolaboratif sesuai  rencana, jalur evakuasi, tiitk kumpul dan display-display yang dibutuhkan sudah terpasang di lokasi secara permanen, lalu sosialisasi dan simulasi untuk memastikan penggunaan display dan langkah-langkah evakuasi sudah di jalankan. Kegiatan ini menghasilkan transfer pengetahuan dan keterampilan kepada siswa-siswa Sekolah Dasar melalui display, sosialisasi dan pelatihan, dimana pengetahuan dan kecakapan siswa tentang bencana meningkat menjadi 100%.