Publication Search

The rapid proliferation of Internet of Things (IoT) devices across diverse industries has significantly increased the vulnerability of IoT edge networks to sophisticated cyber threats. Traditional intrusion detection systems (IDS), such as signature-based and anomaly-based approaches, are often insufficient in addressing the dynamic and evolving nature of these threats. This study proposes a hybrid intrusion detection system (IDS) framework that combines supervised machine learning (ML) techniques with deep reinforcement learning (DRL) to enhance detection performance in real-time, resource-constrained IoT environments. The proposed framework utilizes supervised learning for initial traffic classification and DRL for adaptive decision-making, enabling the system to continuously learn and optimize its detection policies based on new attack patterns. The hybrid approach significantly improves detection accuracy and reduces false positives when compared to conventional signature-based and single-model ML systems. In addition to improved detection capabilities, the framework's computational efficiency allows it to operate effectively within the constraints of IoT devices, ensuring that it is suitable for large-scale deployments. Benchmark evaluations using publicly available datasets, such as NSL-KDD, IoT-23, and BoT-IoT, show that the hybrid IDS framework outperforms traditional methods, providing a more robust and adaptive solution to cybersecurity challenges in IoT edge networks. The findings of this study suggest that combining machine learning with deep reinforcement learning offers a promising approach to secure IoT environments and address the limitations of existing IDS techniques. Future work will explore enhancing real-time adaptability, scalability, and the detection of zero-day attacks in evolving IoT ecosystems.

This study investigates the integration of computational thinking (CT) principles with adaptive curricula to enhance problem-solving skills in undergraduate programming education. Traditional programming curricula often emphasize syntax and basic concepts, neglecting critical problem-solving strategies. The adaptive curriculum framework used in this study combines CT skills such as decomposition, pattern recognition, abstraction, and algorithmic thinking with personalized learning experiences. A mixed-method approach, combining qualitative and quantitative research, was employed to assess the effectiveness of this integrated approach. The results show significant improvements in students' problem-solving abilities, conceptual understanding, and engagement compared to a control group following a traditional curriculum. Students in the experimental group, which received the adaptive curriculum, demonstrated better performance in applying algorithms and debugging code. Additionally, students expressed higher levels of engagement and motivation, suggesting that the personalized learning environment fostered greater academic involvement. The study highlights the importance of integrating CT principles with adaptive learning frameworks to create a more inclusive and effective learning environment that accommodates diverse learning needs. The findings suggest that adaptive curricula can bridge gaps in traditional education by providing personalized support and ensuring that students progress at their own pace. This approach is especially beneficial for programming education, where both conceptual understanding and practical problem-solving skills are critical for success. Future research should explore the long-term impact of adaptive learning frameworks and investigate how these technologies can be integrated with traditional teaching methods to maximize their effectiveness.

User reviews on the Shopee e-commerce platform represent an important source of information for understanding consumer perceptions of products and services. Sentiment analysis is commonly applied to classify user opinions into positive, neutral, and negative sentiment categories based on textual data. This study aims to analyze the performance of the Long Short-Term Memory (LSTM) method in sentiment classification of Shopee user reviews. The dataset used in this study consists of Indonesian-language user reviews that have undergone preprocessing stages, including case folding, text cleaning, tokenization, and stopword removal. The LSTM model was trained using preprocessed text represented as word sequences. Model performance was evaluated using overall accuracy and class-wise classification results. The experimental results indicate that the LSTM method achieved an overall accuracy of 87.62%. In addition, the classification performance for the positive sentiment class reached 95.27%, the neutral class achieved 4.96%, and the negative class reached 74.26%. These results demonstrate that the LSTM method performs well in classifying sentiment in Shopee user reviews, particularly for positive sentiment. This study is expected to provide insights and references for the application of deep learning methods in sentiment analysis of Indonesian e-commerce review data.

The increasing complexity and scale of modern cybersecurity threats necessitate the development of advanced systems capable of efficiently detecting, analyzing, and mitigating incidents in real time. This paper proposes an automated framework for digital forensics and incident response that leverages big data analytics and real time network traffic profiling. The framework integrates cutting-edge technologies, including Apache Spark for real time data processing and Hadoop for scalable data storage, combined with machine learning models like LSTM and Autoencoders to detect anomalies and threats in network traffic. By automating the process of incident detection and response, this framework significantly reduces the time required to identify threats and improves the accuracy of forensic evidence correlation across heterogeneous network environments. The study highlights the advantages of using machine learning models and big data tools to address the limitations of traditional manual and semi-automated systems, which often struggle to keep pace with large-scale data generation. Testing results demonstrate that the proposed framework can handle large data volumes efficiently, providing real time, actionable insights with significantly reduced response times. Additionally, the framework improves forensic analysis by enabling the correlation of evidence from different devices and protocols, making it more effective than traditional methods in identifying the root cause of security incidents. However, challenges related to data heterogeneity, scalability, and system integration were encountered during testing. The proposed framework holds promise for significantly enhancing the efficiency and effectiveness of cybersecurity operations, with future work focusing on further integration of advanced AI techniques and machine learning models for dynamic and adaptive incident response.

Explainable Artificial Intelligence (XAI) has become a critical area of research within artificial intelligence, focusing on improving the transparency and interpretability of machine learning (ML) models, often referred to as "black-box" models. The need for XAI techniques arises from the inherent complexity of ML models, which can make their decision-making processes difficult for users to understand. This study investigates various XAI techniques, including LIME (Local Interpretable Model-agnostic Explanations) and SHAP (SHapley Additive exPlanations), to assess their impact on model interpretability without significantly compromising predictive performance. A comparative experimental design was used, applying these XAI methods to different ML models, including deep neural networks and ensemble methods, within large-scale enterprise data analytics systems. The results indicate that XAI methods significantly enhance model transparency and decision traceability, allowing users to understand the influence of individual features on predictions. While a slight reduction in predictive accuracy was observed, especially with simpler models, the trade-off between interpretability and performance was deemed acceptable, particularly in fields requiring transparency, such as healthcare, finance, and autonomous systems. The use of XAI in enterprise data systems has practical implications for fostering trust and enabling informed decision-making among stakeholders. Furthermore, the study discusses the challenges and limitations of applying XAI techniques, such as complexity, scalability, and model-specific limitations. Future research is suggested to focus on developing more scalable and efficient XAI methods, enhancing their applicability across various model types, and addressing the challenges of real-time applications. This will be crucial in ensuring the widespread adoption of XAI in critical domains, promoting the ethical use of AI while maintaining predictive accuracy.

This study was based on the low ability of third-grade students at SD Negeri 2 Wonorejo to understand syntactic material, particularly the use of conjunctions, due to a lack of interactive and contextual learning media. This study aimed to develop EKUJARSI E-Flipbook learning media based on Kudus local wisdom that has been tested for feasibility and practicality to improve student understanding. The research method used was Research and Development (R&D) with the ADDIE (Analyze, Design, Development, Implementation, Evaluation) development model. The test subjects included subject matter experts, media experts, classroom teachers, and third-grade students. Data collection instruments used validation sheets and user response questionnaires. The results showed that the EKUJARSI media was highly valid for use, as evidenced by a validation percentage of 89.6% from media experts and 86.15% from subject matter experts. The practicality test also showed positive responses, with teachers giving a score of 80% and students giving a score of 85.9%, which is considered very feasible. Specific findings show that the ease of use aspect received a perfect score of 100% from students, indicating that this media is very user-friendly. It is concluded that the EKUJARSI E-Flipbook is feasible and practical to be implemented as an innovative learning media that effectively integrates local cuisine to facilitate understanding of language concepts.

This community service activity aims to enhance students’ literacy through the application of ecoprint as an environmentally based creative learning medium at MAT Daarut Tahfidz Al-Ikhlas Banda Aceh. This program was motivated by the limited utilization of practice-based art activities as a means of strengthening students’ literacy. The implementation method employed a participatory-educative approach consisting of material socialization, ecoprint technique demonstrations, group-based hands-on practice, reflective discussions, and activity evaluation. The results indicate that students were able to understand the concepts and stages of ecoprint, produce works with various natural motifs, and demonstrate improvements in visual literacy and creativity. In addition, the activity increased students’ awareness of the use of natural materials and the importance of environmental friendly practices. Group-based activities also encouraged the development of social literacy through collaboration and communication among participants. Therefore, ecoprint can be considered an effective and contextual alternative learning medium to support the enhancement of students’ literacy in the school environment and has the potential to be sustainably developed.

This study investigates the role of Enterprise Information Systems (EIS) in driving innovation within organizations. The research employs a mixed-method approach, combining survey-based structural analysis and in-depth organizational case studies to explore how different EIS capabilities influence organizational innovation. The study focuses on four key EIS capabilities: functional capabilities such as workforce management and customer value creation; technological capabilities including ERP systems and real-time analytics; dynamic capabilities, especially organizational learning; and collaborative innovation through external partnerships. The survey results reveal that EIS capabilities, particularly data analytics and integration, significantly enhance organizational agility, decision-making, and innovation outcomes. In-depth case studies provide detailed insights into how these capabilities are applied in real-world organizational settings, illustrating their impact on process and service innovation. The findings indicate that the effective integration of EIS across organizational functions, along with improved access to data, contributes to operational efficiency and innovation success. However, challenges such as integration issues, resistance to change, and lack of skilled personnel were also identified as barriers to successful EIS adoption. The study contributes to the literature by offering a comprehensive understanding of how EIS capabilities drive innovation and highlighting the importance of organizational culture and leadership in the adoption process. The research provides practical recommendations for organizations to leverage EIS for fostering innovation, such as focusing on EIS integration, overcoming organizational barriers, and ensuring leadership engagement. Finally, the study suggests future research directions, including the refinement of multi-method approaches and the need for longitudinal studies to better understand the long-term impact of EIS on innovation outcomes.

The increasing adoption of cloud native microservices has brought about significant improvements in scalability, flexibility, and resilience. However, these advancements also introduce substantial security challenges, particularly in distributed environments where traditional perimeter-based security models prove inadequate. This paper proposes a secure architecture for cloud native microservices that integrates Zero trust Network Access (ZTNA) and multi layered encryption techniques to address these security concerns. The architecture operates on the principle of "never trust, always verify," ensuring that access to resources is strictly controlled and continuously monitored. By incorporating multi layered encryption methods such as RSA and AES, the architecture ensures data protection both in transit and at rest, significantly reducing the risk of data breaches and unauthorized access. Through experimental evaluations, the proposed architecture demonstrated its effectiveness in preventing lateral movement, mitigating data leakage, and resisting common attack vectors such as man-in-the-middle (MITM) attacks and privilege escalation. Additionally, the performance of the system remained optimal, with minimal overhead despite the additional security layers. The architecture's scalability and robust security mechanisms make it a viable solution for real-world microservices environments, where both security and performance are crucial. This paper discusses the potential impact of this secure architecture on the broader field of distributed system security and offers recommendations for future work, including the integration of advanced machine learning techniques for real-time threat detection and automated responses, as well as the adaptation of the architecture for emerging technologies like edge computing and 6G networks.

Reading ability is a basic skill that plays a crucial role in the success of students' learning process. However, various studies indicate that the reading ability of junior high school students remains low. This study aims to analyze the reading ability of students at the UPT SMP Negeri 2 Mappakasunggu using Alfred Schutz's social phenomenology perspective. The research approach used was qualitative with descriptive methods. Data collection techniques included in-depth interviews, observation, and documentation of students and teachers. The results indicate that students' low reading ability is not solely caused by cognitive factors but is also influenced by subjective meanings formed through students' social experiences. The lack of a literacy culture in the family and school environment results in reading not being perceived as an important or enjoyable activity. Furthermore, the dominant use of gadgets for entertainment creates habits that reduce students' interest and concentration in reading texts. From Alfred Schutz's social phenomenology perspective, these conditions are related to students' lifeworlds and stock of knowledge, which shape their perspectives and actions toward reading. This study concludes that improving students' reading ability requires a comprehensive approach, taking into account experiences, social interactions, and the formation of meaning in reading in students' daily lives.

This study aims to examine the effect of contextual learning on students’ learning motivation in Indonesian language learning for second-grade students of Madrasah Ibtidaiyah (MI) Salafiyah 1 Kauman. Contextual learning is an instructional approach that connects learning materials with students’ real-life experiences, which is expected to increase their engagement and learning motivation. This study employed a quantitative approach with an experimental research design. The subjects of the study were second-grade MI students divided into an experimental class and a control class. Data were collected through learning motivation questionnaires and classroom observations during the learning process. The collected data were analyzed using statistical techniques to determine differences in learning motivation between the two groups. The results indicate that contextual learning has a positive and significant effect on students’ learning motivation in Indonesian language learning. Students who were taught using contextual learning showed higher learning motivation than those who were taught using conventional learning methods. Therefore, contextual learning can be considered an alternative instructional strategy to enhance students’ learning motivation at the elementary or Madrasah Ibtidaiyah level.

Learning the Qur'an in Islamic Elementary Schools requires an approach that emphasizes not only reading skills but also understanding the meaning and internalizing the Qur'anic values ​​contextually. However, learning practices that are still dominated by conventional methods have the potential to reduce student engagement and interest in learning, especially in narrative materials such as animal stories in the Qur'an. This study aims to examine the effectiveness of the Qur'anic Magic Cards media as a learning medium for animals in the Qur'an in increasing the understanding and interest in learning of students in Islamic Elementary Schools. This study uses a qualitative approach with data collection techniques through observation, interviews, and documentation. The results show that the use of the Qur'anic Magic Cards media can help students understand the meaning of verses more concretely, increase active involvement during learning, and foster interest in learning the Qur'anic material. This media also contributes to creating a fun, contextual, and meaningful learning atmosphere, so that the messages of the Qur'an are not only understood textually but also internalized in the learning process. Thus, the Magic Quran Cards can be used as an innovative and relevant alternative Quranic learning medium for implementation in Islamic Elementary Schools (Madrasah Ibtidaiyah).

This study aims to develop an effective predictive model for identifying students at risk of academic dropout using the Decision Tree and Linear Regression algorithms. The data used are sourced from the public Kaggle dataset Students Dropout and Academic Success, which includes demographic, socioeconomic, and academic performance variables for each semester. The research method includes data preprocessing stages, such as data cleaning, label encoding for categorical variables, numeric feature normalization, and target class adjustment to focus on binary classification, namely Dropout and Graduate. The modeling process is carried out by comparing the performance of the two algorithms using evaluation metrics of accuracy, precision, and recall. The results show that the Decision Tree algorithm has superior performance compared to Linear Regression in mapping non-linear patterns in student data. Feature importance analysis revealed that the number of curricular units in the second semester and tuition payment status are the main predictors of dropout risk. These findings are expected to assist educational institutions in implementing early interventions to improve student academic success.  

This community service activity aimed to strengthen teachers’ competencies in utilizing digital platforms to support project-based learning at a vocational high school. The background of this activity was the limited ability of teachers to manage project-based learning in a structured and systematic manner using digital platforms. The method employed was participatory mentoring, involving needs analysis, activity planning, implementation of digital platform mentoring, monitoring and evaluation, and reflection with follow-up actions. The participants were teachers of SMK Negeri 1 Tukak Sadai, Bangka Barat. The results indicated an improvement in teachers’ understanding and skills in designing, managing, and evaluating project-based learning using digital platforms. Teachers became more confident in integrating technology into learning processes and demonstrated better organization of project activities. This activity contributed to enhancing teachers’ digital competencies and improving the quality of project-based learning practices in vocational education. The findings suggest that participatory mentoring is an effective approach to supporting sustainable digital learning innovation in vocational schools.

This study aims to analyze the implementation of affective learning strategies in Akidah Akhlak subjects to enhance students' self-control at MTs Muhammadiyah 02 Purbalingga. Affective learning strategies focus on character development through the inculcation of religious values such as patience, honesty, and discipline, aimed at helping students manage their emotions, thoughts, and behavior in daily life. The research employed a qualitative method with a descriptive approach. Data were obtained through in-depth interviews with Akidah Akhlak teachers, the principal, and students, as well as classroom observation. Data analysis was conducted using thematic coding to identify patterns in the implementation of affective learning strategies and their impact on students' self-control. The findings indicate that the implementation of affective learning strategies is carried out through five main stages: receiving, responding, valuing, organization, and characterization. Each stage involves structured approaches, such as using inspirational stories from the Qur'an and Hadith, group discussions, value reflections, and case study simulations. These strategies successfully enhanced students' self-control, particularly in emotional regulation, wise decision-making, and consistent behavior aligned with Islamic values. This study concludes that affective learning strategies play a vital role in improving students' self-control. Systematic and collaborative implementation can assist madrasahs in producing a generation that is faithful, knowledgeable, and morally upright, in line with the vision of Islam.

This study examines deep learning as an educational approach within the Indonesian education system by exploring its conceptual foundations, policy frameworks, and practical implementation. The background of the study arises from ongoing challenges in Indonesian classrooms, where teacher-centered instruction and surface learning practices remain dominant, limiting students’ critical thinking, engagement, and real-world application of knowledge. The study aims to analyze how deep learning, characterized by mindful, meaningful, and joyful learning, is conceptualized in educational theory, reflected in national education policies, and implemented in classroom practices. Using a qualitative literature review method, data were collected from peer-reviewed national and international journal articles, academic books, and research reports related to deep learning in education. The data were analyzed through thematic synthesis to identify patterns concerning implementation strategies, learning outcomes, and implementation challenges. The findings indicate that deep learning contributes positively to students’ cognitive development, motivation, engagement, and 21st-century skills, particularly critical thinking, collaboration, and creativity. However, its implementation in Indonesia faces several obstacles, including limited teacher competence, inadequate assessment systems, insufficient contextual learning materials, and unequal technological infrastructure. The study concludes that successful deep learning implementation requires integrated policy support, continuous teacher professional development, contextualized curriculum design, and equitable access to learning resources. These findings provide practical implications for educators, curriculum developers, and policymakers in strengthening sustainable and humanistic education in Indonesia.

This study aims to improve the learning outcomes of fourth-grade students in science through the application of an experimental learning model on the material of changes in the state of matter. The study used a Classroom Action Research (CAR) approach with the Kemmis and McTaggart model which includes the stages of planning, implementation of actions, observation, and reflection. The subjects of the study were 20 fourth-grade students of SDN Daan Mogot 1. The study was conducted in two cycles. Data collection techniques included learning outcome tests, observation of teacher and student activities, and documentation. Data analysis was carried out quantitatively by calculating the average value and percentage of learning completion, and qualitatively through descriptive analysis of the observation results. The results showed that the application of the experimental method was able to improve the quality of the learning process and student learning outcomes. Teacher activity increased from the sufficient category in cycle I to very good in cycle II, while student activity increased from the good category to very good. Increased student activity in observing, discussing, recording results, and drawing conclusions from experiments had a positive impact on understanding the concept of changes in the state of matter. Student learning completion also increased although not all of them reached the classical standard of 80%. Thus, the experimental method is effective in improving the activeness, quality of the learning process, and the science learning outcomes of fourth-grade students. Although material reinforcement and a variety of learning strategies are still needed to optimize learning outcomes.

This study aims to conceptually examine the role of visual media in Arabic vocabulary learning and its relevance in the context of education in the digital era. Vocabulary mastery is a fundamental component in learning Arabic; however, in practice, vocabulary learning often encounters various challenges, such as the abstract nature of vocabulary, low student motivation, and the dominance of conventional teaching methods. This study employed a qualitative approach using library research. Data were obtained through a review of relevant scholarly sources, including books, journal articles, and previous research related to visual media and Arabic language learning. Data were collected through documentation techniques and analyzed using content analysis to examine the concepts, roles, and effectiveness of visual media in Arabic vocabulary instruction. The findings indicate that visual media, both conventional and digital, play a significant role in improving vocabulary comprehension, strengthening memory retention, and increasing students’ motivation and engagement in the learning process. Visual media also help transform abstract vocabulary into more concrete and contextual representations, making learning more meaningful and effective. Nevertheless, the implementation of visual media still faces several challenges, including limited facilities, teachers’ digital competence, and the suboptimal use of technology in learning activities. Therefore, it is necessary to develop innovative instructional strategies and enhance teachers’ competencies to ensure that visual media can be utilized effectively and sustainably in Arabic language learning.

Learning Al-Qur’an and Hadith in secondary schools is still largely influenced by traditional assessment practices that prioritize written examinations and memorization, which are insufficient to capture students’ actual competencies. This situation often leads to less meaningful learning and limits the development of students’ deep understanding and Islamic character. In essence, Al-Qur’an and Hadith education is intended not only to ensure mastery of theoretical content but also to encourage the internalization and application of Islamic values in everyday life. For this reason, authentic assessment is viewed as a suitable approach because it evaluates learning outcomes in a more comprehensive manner, covering cognitive, affective, and psychomotor aspects. This study aims to analyze the implementation of authentic assessment in Al-Qur’an and Hadith learning at the secondary school level and to examine its effectiveness in improving students’ understanding and learning engagement. The research employed a quantitative method, with data collected through a Google Form questionnaire distributed to students and analyzed using descriptive analysis. The results demonstrate that authentic assessment contributes to deeper conceptual understanding, enhances practical skills such as proper Qur’anic recitation based on tajwid rules and hadith memorization, and promotes active, confident, and responsible learning attitudes. Overall, authentic assessment provides more meaningful learning experiences and represents an effective alternative assessment strategy to improve the quality of Al-Qur’an and Hadith learning in secondary schools.

The Muhammadiyah Orphanage in Pamekasan faces major challenges in developing life skills and digital education for its children due to limited facilities, teaching staff, and conventional learning methods. To address these issues, an edutainment-based approach and digital pedagogy intervention were implemented to enhance learning quality, motivation, and preparedness for future social and technological challenges. The activities included workshops and training on Digital Pedagogy and Edutainment learning materials, as well as simulations and role-plays using a Game-Based Learning approach, followed by evaluations and participant plan presentations. The program significantly improved the wards’ digital literacy, particularly in personal security (online safety), digital ethics (cyber ethics), gadget usage, and information management, with the average score rising from 2.84 to 4.10 on a 5-point scale, surpassing the target of 75% of participants in the “good” category. Beyond cognitive aspects, the program also boosted motivation, engagement, communication, problem-solving, and independence. Caregiver training was also provided to ensure program sustainability. It is recommended that the orphanage integrate the Game-Based Learning Digital Safety module into its non-formal curriculum, enhance caregiver capacity through advanced training, and improve IT infrastructure.