Publication Search

Rainfall is a crucial factor in the stability of the Earth's ecosystem and has a significant impact on agriculture, forestry, energy, and water management. However, increasingly unstable climate change makes rainfall patterns difficult to predict accurately using traditional methods. The city of Medan, the capital of North Sumatra Province, has a tropical rainforest climate with an average annual rainfall of approximately ±2200 mm and an average temperature of 27°C. Significant weather fluctuations in this area can trigger flooding when rainfall increases and cause water shortages when rainfall decreases (BMKG, 2021). Therefore, a prediction approach that can manage non-linear and dynamic data is needed. Artificial Neural Networks (ANN) are one of the reliable machine learning methods for detecting data patterns. By using the backpropagation algorithm, the model can gradually reduce prediction errors, making it widely used in weather forecasting applications. In this regard, this study uses ANN with the backpropagation method to forecast monthly rainfall in Medan City by utilizing data from 2022–2024 as training and testing data.

Tawhid education is a fundamental aspect of Islamic education that serves as the foundation for the formation of faith and individual religious character. The family plays a crucial role as the primary educational environment in instilling the values of Tawhid from an early age. However, in modern social realities, the role of families in Tawhid education often faces various challenges, necessitating support from non-formal Islamic educational institutions. This study aims to describe the role of families in Tawhid education through their involvement in the Majlis Taklim ‘Izzatin Nisa Tangerang. This study uses a qualitative approach with descriptive methods. Data collection techniques were carried out through observation, interviews, and documentation of the Majlis Taklim administrators, the congregation, and ongoing learning activities. The results show that families play an active role in Tawhid education through parental role models, the habit of worship at home, and the strengthening of understanding of the faith gained from the Majlis Taklim activities. The Majlis Taklim ‘Izzatin Nisa serves as an educational tool that strengthens the congregation's understanding of Tawhid and encourages its implementation in family life. Thus, the synergy between families and religious study groups has been proven to contribute to strengthening monotheism education and building religious awareness within the community.

The development of academic writing skills among university students remains a serious challenge in higher education, particularly in Indonesian language courses that are often perceived as theoretical and procedural. This study aims to reconstruct Indonesian language learning through an Outcome Based Education (OBE) approach to strengthen students’ scientific writing skills at Universitas Muhammadiyah Maumere. Employing a qualitative research design, data were collected through classroom observations and in-depth interviews with students and lecturers over a three-week period. The findings reveal that OBE based learning encourages clearer learning objectives, increases students’ awareness of academic writing standards, and fosters reflective and critical writing practices. Students demonstrated improved understanding of scientific text structures, argument development, and academic language conventions. Furthermore, the learning process shifted from teacher centered instruction to a more participatory and student oriented model, allowing students to actively construct knowledge through observation, discussion, and writing practice. The study also highlights that the integration of contextual learning and continuous feedback plays a crucial role in enhancing students’ confidence and motivation in academic writing. These findings imply that OBE based Indonesian language instruction can serve as an effective pedagogical strategy to cultivate academic literacy, critical thinking, and responsible scholarly attitudes among university students. The study recommends broader implementation of OBE principles in language education to support sustainable improvement in higher education learning outcomes.

Household plastic waste remains a serious challenge in urban environments, including Jl. Assofa II, Kebon Jeruk, West Jakarta, which is characterized by high population density and intense socio-economic activities. The low level of public awareness in sorting and managing plastic waste, particularly bottle caps, highlights the need for educational approaches from an early age. This community service activity aimed to enhance environmental awareness, creativity, and character development among children through the “Recycled Keychain” project, which involved recycling plastic bottle caps into keychains. The activity was implemented using a Project-Based Learning (PjBL) approach and involved 12 children aged 7–9 years as the main participants. The implementation methods included education on waste management based on the principles of reduce, reuse, and recycle (3R), collection and sorting of bottle caps, and the process of creating creative products. The results showed an increase in children’s environmental awareness and environmentally responsible behavior, the creation of eco-friendly educational products, as well as improvements in creativity, motor skills, and teamwork abilities. In addition, this activity supported the strengthening of the Pancasila Student Profile, particularly in the dimensions of cooperation, critical thinking, creativity, and independence. Overall, the Recycled Keychain project proved to be an effective model of contextual, participatory, and sustainable environmental education in instilling environmental care values from an early age in urban communities.

Stock price prediction remains a complex challenge due to the dynamic and non-linear nature of financial markets, especially for banking stocks like PT Bank Negara Indonesia (Persero) Tbk (BBNI). This study aims to optimize BBNI stock price forecasting by integrating an automated Extract, Transform, Load (ETL) pipeline with the Long Short-Term Memory (LSTM) algorithm within a data engineering framework. Historical data from 2019 to 2025 were processed through a structured ETL sequence—including data cleaning, feature engineering, and MinMaxScaler normalization—to ensure high data quality. The dataset was partitioned into 80% for model training and 20% for testing to ensure rigorous evaluation. The results demonstrate that the systematic ETL approach significantly enhances model stability and predictive accuracy compared to conventional methods. The LSTM model effectively captured long-term temporal dependencies, providing reliable trend forecasts with an impressive test accuracy, achieving a Root Mean Squared Error (RMSE) of 0.0354. This research underscores that integrating robust data engineering practices with deep learning is essential for building resilient financial decision-support systems.

This study aims to evaluate the effectiveness of the Structural Analytic Synthetic (SAS) method in enhancing the early reading skills of first-grade students in Catholic Religious Education at SD YPPK Yohanes Paulus II Obaa, Mappi Regency, South Papua Province. The study employed a Classroom Action Research approach using a cyclical model comprising the stages of planning, implementation, observation, and reflection, conducted over two cycles. The research subjects consisted of 25 students, and data were collected through reading tests, classroom observations, and documentation. The results indicated a significant improvement in students’ early reading skills, with learning mastery increasing from 61% in the first cycle to 100% in the second cycle. This success was associated with the implementation of the SAS method through structured learning stages, progressing from sentences to words, syllables, and letters, accompanied by the use of varied media and intensive teacher guidance. In addition to improving reading skills, the SAS method was also found to enhance students’ interest, participation, and learning motivation. Therefore, the SAS method is recommended as an effective strategy for teaching early reading to young elementary school students, particularly within the context of Catholic Religious Education.

Flood disasters that struck Cikahuripan Village significantly affected the physical, psychological, and learning motivation conditions of elementary school students, particularly at SD N 1 Cikahuripan. In the post-disaster period, students experienced a decline in learning motivation caused by psychological trauma, loss of a sense of security, and limited learning facilities. This Community Service Program (Pengabdian kepada Masyarakat/PkM) aimed to restore students’ psychological conditions while strengthening their learning motivation through an integrated psychosocial and pedagogical approach. The implementation methods included an initial assessment of students’ conditions, trauma healing activities based on play therapy, and the application of Fun Learning methods combined with the “Tree of Dreams” activity to rebuild students’ intrinsic motivation. The program was conducted from 15 to 22 November 2025 and involved 31 elementary school students as well as teachers as sustainability partners. The evaluation results showed significant positive changes, indicated by increased cheerfulness, active participation, confidence in social interaction, and improved learning focus among students. In addition, teachers’ capacity to implement trauma-sensitive teaching practices also improved. This program demonstrates that post-disaster learning motivation recovery requires a holistic approach integrating psychological and academic recovery, and it has the potential to serve as a replicable model for educational interventions in disaster-prone areas.

This study is motivated by the increasing need for accurate modeling and classification of one-dimensional signal data in intelligent systems. The rapid development of deep learning has led to the adoption of more adaptive and complex neural network architectures capable of capturing both temporal dependencies and local patterns in sequential data. This research aims to analyze and compare the performance of several deep learning models, namely Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid Convolutional Neural Network–GRU (CNN–GRU) model for signal data classification. The research method employs a quantitative experimental approach involving data preprocessing, windowing, model training, and performance evaluation. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. The results indicate that the hybrid CNN–GRU model outperforms the other models, particularly in capturing local features and long-term temporal dependencies within signal data. These findings suggest that the integration of convolutional layers and recurrent mechanisms enhances feature representation and learning stability. This study is expected to contribute both theoretically and practically to the development of deep learning models for signal processing and time-series-based intelligent applications.

This research is based on the problem of declining morals of third-grade students at SD IT Baitul Muslim, which has implications for daily behavior, particularly in Pancasila Education learning. Students tend to exhibit attitudes that are inconsistent with positive character values, such as a lack of tolerance and a low sense of love for the homeland. The purpose of this study is to identify the significant influence of educators' roles in shaping the character of third-grade elementary school students in Pancasila Education subjects. The method used is an ex post facto method with a quantitative approach and a purposive sampling design. The study population consisted of 10 educators and 95 students, with a sample of 10 educators and 40 students from two selected classes. Data collection techniques were carried out through questionnaires and documentation that had been tested for validity using Aiken's V and reliability with Cronbach Alpha. Hypothesis testing using simple linear regression analysis showed a significance value of 0.044 which is smaller than 0.05 and a calculated F value of 5.692 which is greater than the F table of 2.27. Thus, the null hypothesis is rejected and the alternative hypothesis is accepted. The research findings confirm the significant influence of educators, particularly as demonstrators, classroom managers, facilitators, and evaluators, in shaping students' character in Pancasila Education. The implication of these findings is that educators need to optimize role models to improve students' character development.

The purpose of this classroom action research project is to find out how well seventh-grade female students at Mutiara Qur'an Islamic Boarding School in Wonogiri learn Arabic with the use of the audiolingual approach. By emphasizing active listening and pronunciation, the audiolingual approach makes it easier for students to comprehend and communicate in Arabic. Twelve female seventh-graders at Mutiara Qur'an Islamic Boarding School participated in the researcher's qualitative study in order to gather the research findings. The study was conducted in two cycles to enhance Arabic language learning. Planning, action, observation, and reflection are the four components of each cycle in the Kurt Lewin model. Tests, observations, and interviews were the methods utilized to get the data. The proportion of learning mastery and descriptive analysis were used to examine the collected data. The study's findings show that the audiolingual approach is successful in raising female students' Arabic language learning outcomes.

This study examines the development of a Crude Palm Oil (CPO) price forecasting model using an artificial neural network algorithm, specifically the backpropagation algorithm. As one of Indonesia’s main export commodities, CPO has a significant economic impact and influences the income of oil palm farmers. The CPO price data used in this study were obtained from CIF Rotterdam, covering the period from January 2019 to December 2023. The research methodology consists of several stages, including data collection, preprocessing, model design, and model implementation using Python programming. The training results of the backpropagation algorithm show an error value of 0.537829578 after 1,000 epochs, while the evaluation using Mean Squared Error (MSE) indicates an MSE of 0.022709 during the training process and 0.017604 during the testing process. The model also produces CPO price predictions for the next three months, namely 932.578 for the first month, 949.568 for the second month, and 774.855 for the third month. These findings indicate that the developed model is capable of predicting future CPO prices with adequate accuracy, which can assist companies in making better financial decisions and managing risks associated with CPO price fluctuations.

The ability to analyze ethical and health law issues is one of the learning outcomes that must be mastered by students of the Public Health Study Program. Over the past two years, the implementation of lectures has been predominantly based on problem-based learning; however, most student learning outcomes have remained in the “not yet achieved” category, including in completing assignments related to ethical and health law problem solving. This study aims to compare students’ problem-solving abilities between the implementation of project-based learning and problem-based learning. The research was conducted in the Public Health Study Program at Universitas Jambi involving first-semester students enrolled in the Ethics and Health Law course. The research employed a quasi-experimental design involving 200 students, with 100 students assigned to the experimental group and 100 students to the control group. The instruments used were assessment sheets and observation sheets. Data collection was carried out twice, before and after the implementation of the two learning models. Data processing and analysis were conducted using an independent t-test.The results of the study indicate an improvement in students’ problem-solving abilities through the implementation of project-based learning. Project-based learning contributes to strengthening students’ perseverance and problem-solving skills and is therefore recommended for implementation in all courses at higher education institutions.

This study aims to explore the dynamics of psychological empowerment among women entrepreneurs by examining how they construct entrepreneurial identity, develop competence, cultivate independence, and generate social impact through their business activities. The background of this research stems from the need for women in micro and small enterprises to receive support that extends beyond technical skill development, emphasizing psychological aspects such as self-belief, personal control, and decision-making capacity. Employing a qualitative phenomenological approach, this study involved three women entrepreneurs selected through purposive sampling. Data were collected using semi-structured interviews to capture the participants’ subjective experiences in managing their businesses. The findings reveal four central themes: entrepreneurial identity shaped by personal values and lived experiences, competence that evolves through continuous learning and problem-solving, independence reflected in decision-making and business management, and social impact manifested in the creation of economic opportunities and contributions to the surrounding community. These results illustrate that psychological empowerment among women entrepreneurs is a dynamic process formed through the interaction of personal experiences, entrepreneurial practices, and community support. The study emphasizes the importance of empowerment programs that integrate psychological strengthening alongside technical training, with communities serving as vital spaces for collective learning and mutual support. Recommendations highlight the need for holistic empowerment models that consider the sociocultural context of women entrepreneurs and foster sustainable growth in both personal and business capacities.  

The retail industry generates an extremely large and continuously growing volume of transactional data along with the advancement of digital technology, thereby requiring sophisticated and systematic data analysis approaches to support effective and evidence-based business decision-making. This study aims to analyze retail sales data by utilizing the Retail Sales Dataset obtained from the Kaggle platform, which consists of 100,000 transaction records and broadly represents the characteristics of retail transactions. The main focus of this study is to classify product categories and predict customer segments, including the identification of high-spending customers (high spenders), based on demographic attributes such as age and gender, as well as various transaction-related features. The research methodology includes data preprocessing, label encoding, and feature engineering to generate additional variables, including Age_Group, Is_Holiday, and Spender_Group, which are expected to enhance the predictive capability of the models. Several machine learning algorithms, namely Decision Tree, Random Forest, and XGBoost, were implemented and evaluated to compare their respective performance. The experimental results indicate that multiclass product category classification achieves relatively low accuracy, ranging from 27% to 34%. These findings suggest the high complexity of retail data and highlight the need for further model optimization, class balancing techniques, and feature refinement to improve predictive performance in future studies.

The waste problem is still an environmental issue that requires serious attention, especially regarding the low level of public awareness in sorting organic and inorganic waste at its source. Environmental education needs to be given from an early age to form attitudes and behavior that care for the environment in a sustainable manner. This community service activity aims to increase the understanding and awareness of grade III–IV elementary school children towards sorting organic and inorganic waste through a game-based learning approach. The activity was carried out in the Bojongsari area, Depok, involving 10 children as participants. The methods used include delivering interactive material, waste sorting competitions through the Eco-Race Little game, and making mini posters with environmental themes. The results of the activity showed that around 90% of participants were able to differentiate and sort organic and inorganic waste correctly. Apart from increasing cognitive understanding, this activity also fosters an attitude of cooperation, creativity and concern for the environment. Thus, the Little Eco-Race activity is considered effective as a fun and applicable environmental education model for elementary school age children.

This study aims to analyze the obstacles in Social Studies (IPS) learning related to global issues influenced by technological advancement through a literature review method. Social Studies plays an essential role in shaping students’ understanding of social, economic, cultural, and political life while fostering global awareness in the digital era. However, in practice, IPS learning often faces various challenges that reduce its effectiveness and attractiveness. One of the main problems is students ’low interestin IPS, as it is often perceived as boring and overly theoretical, combined with the influence of technology that diverts students’ attention toward digital entertainment. Other contributing factors include non-interactive teaching methods, limited learning facilities, unconducive classroom environments, and a lack of parental support. Through a comprehensive literature review, this study found that relevant strategies to overcome these issues include Project-Based Learning, Technology Integration, and Contextual Approaches. These strategies enable students to learn actively, critically, and collaboratively by linking IPS material to global issues and real-world phenomena. The recommended solutions include implementing project-based education, utilizing interactive digital media, and fostering intercultural dialogue within the learning process. The results of this study are expected to provide theoretical contributions to the development of adaptive IPS learning models in response to technological progress, as well as practical benefits for teachers, schools, and policymakers in creating Social Studies learning that is engaging, meaningful, and relevant to the global challenges of the 21st century.

This study investigates how faith-based giving can be transformed into measurable development impact in the education sector, focusing on the Global Muslim Philanthropy Fund for Children (GMPFC) established by the Islamic Development Bank (IsDB) in partnership with UNICEF. Traditional Islamic philanthropic instruments, such as zakat and sadaqah, often provide short-term relief but lack structured governance, limiting their long-term impact on educational outcomes. Using a qualitative-explorative, this research analyzes secondary data from 2021–2025, including institutional reports, program documents, and peer-reviewed literature, to assess how GMPFC operationalizes faith-based resources through pooled, multilateral, and impact-oriented mechanisms. The findings indicate that GMPFC strategically funds education-enabling conditions, including child health, nutrition, psychosocial wellbeing, and youth empowerment, which are empirically linked to school readiness, retention, and learning quality. Comparative analysis shows that GMPFC outperforms traditional philanthropy and conventional aid by combining cultural legitimacy, institutional rigor, and alignment with Sustainable Development Goals (SDGs). Its governance model ensures standardized monitoring, fiduciary oversight, and cross-sectoral integration, addressing longstanding limitations of fragmented philanthropic delivery. Despite its effectiveness, the study highlights a lack of longitudinal learning outcome data, limiting precise quantification of educational impact beyond enabling conditions. Nonetheless, GMPFC exemplifies a hybrid development-finance model, demonstrating how Islamic philanthropic values can be operationalized to generate sustainable, measurable contributions to child education and human capital formation. These findings offer actionable insights for policymakers, development practitioners, and faith-based organizations aiming to scale philanthropic resources for education in vulnerable contexts.

This study aims to: (1) describe the implementation of local wisdom–based learning in the teaching process of Grade VII B at UPT SMP Negeri 4 Kelara Jeneponto; (2) determine the improvement of students’ social awareness after the implementation of local wisdom–based learning; and (3) identify the enhancement of students’ learning culture through the application of this learning model. The research employed a Classroom Action Research (CAR) design using the Kemmis and McTaggart model, conducted in two cycles consisting of planning, action, observation, and reflection. The subjects were 25 seventh-grade students, focusing on two main variables: social awareness and learning culture. Data were collected through observation, questionnaires, and documentation, and analyzed using descriptive quantitative and qualitative techniques. The findings indicate that the integration of Jeneponto cultural values Tabe’ Karaeng, Sipakatau, Sipakalabbiri, Sipakainga, Siri’ na Pacce, Akbulo Sibatang, and empo sipitangari was implemented effectively and became more optimal in Cycle II after improvements were made through Problem-Based Learning (PBL) and the use of more contextual local cases. Students’ social awareness increased from an average of 63.13% in Cycle I (categorized as Fair) to 80.42% in Cycle II (categorized as High). Similarly, students’ learning culture improved from 69.69% (Fair) in Cycle I to 82.66% (High) in Cycle II, meeting the classical mastery criteria. The study concludes that integrating local wisdom into classroom instruction not only strengthens students’ understanding of social studies content but is also effective in shaping their social character and learning habits. This model is recommended for development in other subjects and educational levels.

The policy to implement the deep learning approach in Indonesian schools starting from the 2025/2026 academic year requires teachers to be well prepared to design and conduct meaningful and in-depth learning. This community service program aimed to enhance the understanding and pedagogical competence of senior high school and madrasah economics teachers in Tasikmalaya City in implementing the deep learning approach comprehensively. The program employed a community-based science and technology application scheme through training and mentoring activities conducted in face-to-face, online, and hybrid modes, consisting of planning, implementation, and evaluation stages. The training materials covered the concept and implementation of deep learning, lesson planning, instructional design and models, learning evaluation, development of innovative technology-based learning media, and reinforcement of microeconomic and macroeconomic content. A total of 78 economics teachers participating in the MGMP of Tasikmalaya City were involved in the program. The results indicate that the training improved teachers’ understanding of deep learning, lesson planning, and assessment strategies that support meaningful learning. Participants’ reflections revealed high motivation and strong relevance of the training materials to their professional duties. This program contributes to strengthening teachers’ readiness to implement deep learning sustainably and supports the improvement of economics learning quality in schools.

Because it directly impacts revenue, customer loyalty, and long-term business sustainability, customer churn is a critical issue for the e-commerce industry. High churn rates indicate that a business is unable to retain existing customers, which means it is more expensive to acquire new customers. Therefore, a precise analytical approach is needed to identify customer behavior patterns that are likely to churn. Using machine learning methods, this study analyzes and predicts customer churn. For this study, the E-Commerce Customer Churn 2025 dataset, obtained from Kaggle, was used. This dataset consists of 10,000 customer data and contains fifteen variables covering transaction behavior, customer characteristics, and churn status. Data preprocessing, descriptive analysis, exploratory data analysis (EDA), and classification model development using Logistic Regression and Random Forest algorithms were part of the research project. Model evaluation was conducted using a Confusion Matrix and Receiver Operating Characteristic (ROC) Curve to evaluate the model's accuracy and ability to distinguish between churned and non-churned customers. The results showed that the Random Forest model performed better than Logistic Regression, with an ROC-AUC of 1.00. Furthermore, feature importance analysis revealed that the days_since_last_purchase variable was the most dominant factor in predicting customer churn. These findings are expected to help e-commerce companies design more effective, data-driven customer retention strategies.