Publication Search

Student dropout in higher education remains a persistent socioeconomic challenge, yet many predictive models reported in the literature are methodologically compromised by randomized cross-validation schemes that introduce temporal data leakage and artificially inflate predictive performance. This study proposes a longitudinal prescriptive learning analytics framework integrating three complementary methodological components: a Leave-One-Cohort-Out (LOCO) temporal validation protocol, a hybrid SMOTE-ENN class balancing strategy, and temporal velocity feature engineering derived from Learning Management System (LMS) behavioral trajectories. The framework was evaluated on a longitudinal dataset comprising 464,739 enrollment records and 77 features. Five predictive algorithms—XGBoost, LightGBM, CatBoost, Random Forest, and Logistic Regression—were comparatively assessed on a strictly isolated blind holdout cohort (2022), with CatBoost emerging as the champion estimator, achieving a PR-AUC of 0.8859, a Macro F1-Score of 0.9143, and the lowest Brier Score (0.0221), thereby demonstrating superior calibration and discriminative capability under severe class imbalance (93:7 ratio). Comprehensive ablation analysis revealed that temporal velocity features function not merely as additive predictors, but as a structural prerequisite enabling Synthetic Minority Oversampling Technique with Edited Nearest Neighbors (SMOTE-ENN) to generate high-quality synthetic boundary instances; removing these features reduced minority-class precision from 0.8302 to 0.6721. To operationalize predictive outputs into actionable intervention pathways, Diverse Counterfactual Explanations (DiCE) were implemented under a three-tier causal constraint architecture on 96 borderline high-risk students, generating 384 feasible intervention scenarios exclusively targeting forward-looking behavioral velocity metrics without constraint violations. Collectively, these findings advance the paradigm of prescriptive learning analytics by providing educational institutions with interpretable risk diagnostics and operationally feasible intervention guidance grounded in empirically validated behavioral and temporal dynamics.

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.

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 develops an interpretable machine learning model to predict the settlement status of Hajj fees in Jambi Province, Indonesia. Utilizing the XGBoost algorithm on a dataset of 4,332 prospective pilgrims from 2025, the research addresses the critical challenge of class imbalance where only 28.5% of samples are labeled "Unsettled". The baseline XGBoost model achieved a ROC-AUC of 0.7778, with a recall of 0.3482 for the minority class. SHAP (SHapley Additive exPlanations) analysis was employed to interpret model predictions, revealing that financial features specifically NILAI_VA (Virtual Account Value), JML_SETORAN (Deposit Amount), and JML_PELUNASAN (Settlement Amount) are the most significant factors influencing repayment risk, with negative SHAP values indicating increased default probability. The findings demonstrate that an interpretable XGBoost framework can provide both predictive accuracy and actionable insights for policymakers, enabling targeted interventions such as flexible payment schemes and enhanced financial monitoring for high-risk pilgrims..

Hypertension is a major global health risk that requires accurate early detection, yet conventional methods struggle with complex and imbalanced health datasets. This study aims to optimize hypertension prediction using a Logistic Regression model integrated with Borderline-SMOTE to enhance recall and provide model transparency through SHAP (Shapley Additive Explanations). The method utilizes the BRFSS dataset, applying Borderline-SMOTE to address class imbalance at the decision boundary and XAI techniques for global and local interpretation. The findings show that the model achieved an accuracy of 0.719, an AUC of 0.800, and a significantly improved recall of 0.756. SHAP analysis identified age, high cholesterol, and BMI as the most influential risk factors, while waterfall plots successfully clarified individual risk extremes, ranging from 1.72% to 99.43% probability. These results imply that the proposed approach provides a sensitive and transparent screening tool for public health practitioners, effectively balancing statistical efficiency with clinical accountability.

This study aims to analyze and project consumer prices of cabbage commodities at four levels: Ngawi Regency, Pacitan Regency, East Java Province, and nationally, using the additive Holt–Winters forecasting model. Monthly price data for the period January 2020–December 2024 were used to capture the dynamics of levels, trends, and seasonal patterns that affect price fluctuations. Model performance was evaluated using the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) indicators. The results showed differences in model accuracy between regions. East Java Province produced the best performance with the lowest MAE and RMSE values, indicating a more stable price pattern that was easier for the model to capture. In contrast, Ngawi Regency showed the highest volatility, resulting in greater forecasting errors. Pacitan Regency displayed a relatively consistent seasonal pattern with moderate accuracy, while national data showed smoother fluctuations due to the aggregation effect. Overall, the additive Holt–Winters model is effective for short-term projections in regions with low to moderate variability, but is less optimal in regions with highly volatile price dynamics.

This study aims to examine the use of cattle waste as organic fertilizer to minimize agricultural operational costs in Mrawan Village, Tapen District, Bondowoso Regency. Cattle waste, particularly manure, is an abundant local resource that has not been optimally utilized by the local community. The majority of farmers in the village still rely on chemical fertilizers, which are relatively expensive and have a negative impact on long-term soil health. Therefore, this program is designed to provide a sustainable alternative solution through an educational approach and community empowerment. The methods used in this activity include outreach, technical training, and direct assistance in the process of making organic fertilizer from cow manure. Education focuses on simple fermentation techniques, the composition of natural additives, and appropriate fertilizer application methods. Farmers are actively involved in every stage of the activity, so they become not only beneficiaries but also agents of change in environmentally friendly agricultural practices. The results of the activity indicate that the use of organic fertilizer from cattle waste can reduce the cost of purchasing chemical fertilizers by up to 40% in a single planting season. In addition, organic fertilizer has been shown to increase soil fertility, improve soil structure, and support healthier plant growth. Environmental impacts are also reduced, as livestock waste management is more controlled and does not pollute water or air sources. Therefore, utilizing cattle waste as organic fertilizer not only reduces environmental pollution but also provides an economic and ecological solution that benefits local farmers. This program is expected to become a model for empowerment that can be replicated in other areas with similar characteristics.

The selection system at Bina Negara Gubug Vocational School Jl. KH. Hasan Anwar No.9 Gubug currently processes data on the criteria for each student for each type of scholarship. It does not yet have a database system but uses a computerized system with Microsoft Excel, so there are often delays in the selection process in preparing the selection report for scholarship recipients. This research uses the Research and Development (R & D) development model by Borg and Gall with 6 steps of development, namely Research and Information Collecting, Planning, Develop Premilinary Form of Product, Premilinary Field Testing, Main Product Revision, Main Field Testing. The scholarship selection decision support system application product uses the SAW (Simple Additive Waighting) method. Visual Basic 6.0 development software and Microsoft Access database. This system can provide a useful solution for the decision-making system for selecting scholarship recipients for schools so that a better and faster selection can be achieved.      

Bali is a renowned tourist destination that attracts visitors from around the world, particularly for its natural beauty, rich culture, and delicious cuisine. The increasing number of tourists in Bali has driven rapid growth in the culinary industry. In Denpasar City, selecting the right location is a key factor for the success of culinary businesses, as each location has different characteristics and potentials. This study employs the Multiple Attribute Decision Making (MADM) model, combining the Simple Additive Weighting (SAW) and Technique for Orders Preference by Similarity to Ideal Solution (TOPSIS) methods, to determine the optimal location for culinary businesses in Denpasar City. Data were collected through surveys of 154 culinary business owners, considering eight criteria: Accessibility, Visibility, Traffic, Facilities, Expansion, Environment, Competition, and Regulations. The study's findings indicate that both SAW and TOPSIS methods identify high population density areas as the best choice. The SAW and TOPSIS method provides the highest preference value of 0,8815 and 0.7082 respectively, making it the more effective method for recommending optimal culinary locations in Denpasar City.

Information systems now play a crucial role in various fields, including education. Yayasan Aldiana Nusantara (YAN), which supports students from low-income backgrounds, faces challenges in selecting scholarship recipients. To streamline this process, this research designs a web-based Decision Support System (DSS) using the Simple Additive Weighting (SAW) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) methods. The system considers criteria such as class, KPS/KIP/KKS recipients, parents' income, report card grades, and extracurricular activities. The DSS is developed using the Waterfall model, PHP, and MySQL, with four types of access rights: Administrator, Judge, Student, and Foundation Leader. Sensitivity testing shows that the SAW method is more responsive to changes in criteria compared to TOPSIS, with a sensitivity value of 2.98 for SAW and 0.022 for TOPSIS. These results indicate that SAW is more optimal in assisting YAN in effectively and efficiently selecting scholarship recipients.

Background: Additive manufacturing (AM) requires reliable and efficient defect detection mechanisms to ensure structural integrity and product quality, yet conventional inspection approaches remain time-consuming and often unsuitable for real-time industrial deployment. Objective: This study aims to develop and experimentally validate an artificial intelligence based vision inspection system capable of accurately detecting surface defects in AM components. Methods: A Convolutional Neural Network (CNN) architecture utilizing pretrained backbones (ResNet and EfficientNet) was implemented with a transfer learning strategy and data augmentation techniques. High-resolution AM surface images representing porosity, cracks, and layer misalignment were used for training and evaluation. Model performance was assessed using Accuracy, Precision, Recall, F1-score, and mean Average Precision (mAP), and comparative benchmarking was conducted against traditional machine learning models such as Support Vector Machine and Random Forest. Results: The proposed CNN-based models significantly outperformed conventional approaches, achieving up to 95.1% Accuracy and 92.8% mAP. The EfficientNet backbone demonstrated superior generalization capability, particularly in balancing Precision and Recall, indicating robust defect detection performance across multiple categories. These findings confirm that AI-driven inspection frameworks provide scalable and reliable quality assurance solutions for advanced manufacturing environments.

Autonomous vehicles (AVs) are revolutionizing transportation by relying on advanced AI techniques like deep learning and reinforcement learning for decision-making and navigation. However, concerns about the opacity of traditional AI models in safety-critical applications such as autonomous driving raise issues related to safety, accountability, and trust. This study explores the integration of Explainable AI (XAI) techniques in AV systems to enhance transparency and interpretability while maintaining high prediction accuracy. XAI methods, such as LIME (Local Interpretable Model-agnostic Explanations) and SHAP (SHapley Additive ExPlanations), provide understandable justifications for AI-driven decisions, addressing biases, fairness, and accountability. These techniques also support regulatory compliance and foster public trust in AVs. A mixed-methods approach, combining experimental simulations and user surveys, was employed to integrate XAI into AV systems and test its performance in urban traffic and highway driving scenarios. Feedback from users, collected through questionnaires and in-depth interviews, revealed that XAI-enhanced systems significantly improved the interpretability of AV decisions, leading to higher user trust and satisfaction. The study highlights the importance of balancing model complexity with interpretability, demonstrating that XAI techniques are crucial for building trust and ensuring accountability in autonomous driving systems.

Artificial food additives pose significant health risks to Gulf Cooperation Council (GCC) citizens despite regional authorities' extensive medical, legislative, and technological efforts. Literature highlights the detrimental impacts of these additives, including malnutrition, digestive disorders, respiratory problems, skin issues, hives, nausea, diarrhea, shortness of breath, allergic reactions, high blood pressure, and tumors. The research project at hand aims at becoming the first official and comprehensive mobile application of its own in the GCC region that manages the calculation and demonstration of an up-to-date health and legal knowledge base of the impacts of artificial additives, enhances the awareness, automatically recognizes the artificial additives, and provides alternative solutions, for both android and IOS mobile platforms. This research project introduces "Weqaya," a pioneering mobile application designed to manage, educate, and raise awareness about the effects of artificial additives. Weqaya provides real-time health and legal information, identifies additives, and suggests alternative solutions for Android and iOS platforms. The project employs an Agile-based SDLC model to explore, develop, and evaluate the food additive phenomena in Weqaya. The application's usability evaluation scores a promising 95.21%, indicating its potential utility for GCC health ministries, dietitians, academics, researchers, and food producers in enhancing knowledge and promoting non-artificial food options.

Employees are Human Resources who are very important to achieve the company's vision and mission. Employees have a significant impact on the company's growth and ability to compete in the market. One way to create quality human resources, we need a system to determine exemplary employees. One way to determine employees by using a decision support system. Decision support systems are made to determine exemplary employees in a company based on predetermined weights. The method used in this system is Simple Additive Weighting because it has the advantage of being able to determine the weight value of each attribute. The results obtained from calculations using this system are the names of exemplary employees who have been sorted according to their ranking from the top to the bottom and this system can be used by companies to make it easier to give rewards or appreciation in determining exemplary employees.

Curah hujan adalah salah satu faktor penting pada banyak sektor, seperti salah satunya yaitu pada bidang perkebunan. Tinggi-rendahnya curah hujan mempengaruhi proses perawatan hingga masa panen dalam sektor ini. Maka itu butuh mengketahui kapan saat yang benar untuk melakukan perawatan agar tidak mengalami kerugian. Diperlukan peramalan untuk menentukan waktu yang tepat. Analisis yang dipergunakan pada observasi ini merupakan Holt-Winters Exponential Smoothing. Hasil observasi menggunakan model terbaik ditentukan dengan besarnya nilai MAPE, dengan model Additive () menunjukkan nilai MAPE sebesar 0,3271700 sebagai model yang terbaik, MAPE dapat menjadi alat ukur untuk menentukan model terbaik, karena jika nilai MAPE semakin kecil maka semakin baik pula sebuah model. Dari hasil peramalan tersebut dapat diketahui bahwa intensitas hujan yang lemah berada saat bulan juni, juli hingga september disetiap tahunnya, pada saat itu para petani dapat melakukan perawatan pada tanaman sawit.