Publication Search

Demam berdarah dengue merupakan tantangan kesehatan masyarakat yang terus berulang di wilayah tropis, termasuk Indonesia. Penelitian ini bertujuan untuk memprediksi jumlah kasus tahunan dengan memanfaatkan lima algoritma pembelajaran mesin, yaitu Regresi Linier, Decision Tree, Random Forest, Support Vector Machine (SVM), dan Neural Network. Data historis tahun 2017–2024 diolah menggunakan teknik windowing deret waktu untuk menghasilkan fitur lag yang sesuai bagi pembelajaran terawasi. Evaluasi kinerja dilakukan melalui metrik Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), serta koefisien determinasi (R²). Model Decision Tree menunjukkan performa paling unggul pada sebagian besar indikator. Prediksi untuk tahun 2025 mengindikasikan adanya peningkatan moderat jumlah kasus. Namun, rendahnya nilai R² pada seluruh model mengisyaratkan perlunya pendekatan multivariat yang lebih kompleks dengan mempertimbangkan faktor iklim, lingkungan, dan demografi. Hasil penelitian ini menegaskan pentingnya kualitas data dan pemilihan fitur yang tepat dalam peramalan epidemiologis guna mendukung perencanaan kesehatan yang lebih efektif.

Background: The development of industrial technology in the Industry 4.0 era has encouraged the implementation of intelligent monitoring systems to improve machine reliability and operational efficiency. However, machine fault diagnosis systems based on artificial intelligence often face limitations in terms of interpretability because the models used are complex and difficult to explain. Objective: This study aims to develop a deep learning-based industrial machine fault diagnosis system integrated with an Explainable Artificial Intelligence (XAI) approach to improve diagnostic accuracy while providing interpretable insights for users. Method: The research method involves collecting data from industrial machine sensors consisting of vibration signals, temperature measurements, and acoustic signals, followed by data preprocessing and feature extraction processes. The processed data are then used to train a deep learning-based diagnostic model, after which explainability methods such as SHAP or LIME are applied to analyze the contribution of each feature to the model’s prediction results. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Results: The results indicate that the proposed deep learning model achieves better performance compared to conventional machine learning methods such as Support Vector Machine and Random Forest. Furthermore, the explainability analysis reveals that vibration amplitude, increases in machine component temperature, and anomalies in acoustic signals are the main factors influencing machine fault detection. Therefore, the proposed system not only improves the accuracy of machine fault diagnosis but also provides transparency in the decision-making process, thereby supporting the implementation of predictive maintenance in smart manufacturing environments.

Mining operations rely heavily on the performance and reliability of heavy equipment used in the production process. One of the most important hauling units in open-pit mining is the dump truck, which functions to transport overburden and coal from the mining front to disposal areas. Due to high operational intensity, dump trucks require effective maintenance management to ensure equipment reliability and reduce unexpected downtime. However, maintenance activities are often carried out based only on routine service schedules without analytical planning based on historical data. This study aims to analyze the implementation of forecasting methods in maintenance management to improve the effectiveness of dump truck maintenance planning in mining operations. The research was conducted during field work practice at PT Putra Perkasa Abadi Jobsite BIB, Tanah Bumbu, South Kalimantan. The data used were historical maintenance records of dump truck units obtained from the maintenance department. The research method used a quantitative approach with time series forecasting analysis to identify maintenance patterns and estimate future maintenance needs. The results show that forecasting-based maintenance planning can help companies predict maintenance requirements more accurately and prepare maintenance resources more efficiently. Furthermore, the implementation of forecasting methods can reduce unexpected equipment failures and support operational efficiency in mining activities.

The integration of big data in the financial sector has increasingly attracted scholarly attention, particularly in areas such as risk management, fraud detection, algorithmic trading, and investment optimization. Given the rapid development of this field, it is essential to map research trends and identify emerging directions that shape the future of financial innovation. This study applies a bibliometric approach using 3,829 articles retrieved from the Scopus database from 1981 to 2025, with data processed through R Studio and the Bibliometrix-Biblioshiny application. The objective is to explore the intellectual landscape of big data finance and reveal research frontiers as well as thematic evolution. The results show a sharp increase in publications after 2015, alongside the growth of fintech and artificial intelligence applications, with dominant themes including blockchain integration, risk analytics, and predictive modelling. Cross-disciplinary and cross-regional collaborations continue to expand. These findings provide a comprehensive overview of how big data has shaped financial studies and offer insights for potential future research directions.

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..

This study aims to analyze the factors affecting real wages of construction workers across provinces in Indonesia from 2010 to 2023 using panel data analysis. The independent variables include Provincial Minimum Wage (UMP), Consumer Price Index (CPI), Open Unemployment Rate (TPT), and Performance Pay (Balas Jasa). A panel dataset of 476 observations from 34 provinces over 14 years was analyzed using three model approaches: Common Effect Model (CEM), Fixed Effect Model (FEM), and Random Effect Model (REM). The best model was determined through Chow Test, Hausman Test, and Lagrange Multiplier Test, which confirmed that the Fixed Effect Model (FEM) is the most appropriate for analyzing this research data. FEM estimation results show that simultneously, all independent variables (UMP, CPI, TPT, and Performance Pay) have a significant effect on real wages with an F-statistic value of 436,465.9 (p-value = 0.0000 < 0.05), indicating that the model as a whole is highly valid and capable of explaining the variation in real wages collectively. However, partial tests reveal that only the Real Wage variable has a positive and statistically significant effect on Performance Pay (coefficient = 106.3320; t-statistic = 1276.083; p-value = 0.0000), while UMP (p-value = 0.1472), CPI (p-value = 0.6460), and TPT (p-value = 0.6934) show no significant effects at the 5% significance level. The research model demonstrates very high predictive ability with an R-squared value of 0.999735 (99.97%), indicating that the variables studied can explain nearly all variation in real wages of construction workers at the provincial level. This research provides policy implications that improving real wages in the construction sector requires an integrated approach that focuses not only on minimum wage setting but also on regional inflation control, human capital quality improvement, and creating conducive labor market conditions through unemployment reduction

End-to-end autonomous driving has emerged as a promising paradigm in which deep neural networks directly map raw visual inputs to continuous control actions. Despite its effectiveness, this approach suffers from limited transparency, posing significant challenges for deployment in safety-critical driving scenarios. This study addresses the lack of interpretability in vision-based end-to-end autonomous driving systems and aims to analyze model decision-making behavior under critical conditions such as sharp steering maneuvers and abrupt control transitions. To this end, an explainable end-to-end autonomous driving framework is proposed, combining a convolutional neural network trained via imitation learning with gradient-based visual attribution techniques, including Grad-CAM. The model predicts continuous steering, throttle, and braking commands directly from front-facing camera images, while explainability mechanisms are applied to reveal input regions influencing each control decision. Model performance is evaluated using both prediction accuracy and safety-oriented behavioral metrics. Experimental results show that the proposed explainable model achieves lower control prediction errors compared to a baseline end-to-end CNN, reducing steering mean squared error from 0.034 to 0.031, throttle error from 0.021 to 0.019, and brake error from 0.018 to 0.016. Moreover, safety-oriented analysis indicates improved driving stability, with steering variance reduced from 0.087 to 0.072 and abrupt control changes decreased from 14.6 to 10.3 events. Visual explanations consistently highlight road surfaces and lane-related structures during complex maneuvers, indicating reliance on semantically meaningful cues. In conclusion, the results demonstrate that integrating explainability into end-to-end autonomous driving not only preserves predictive performance but also correlates with smoother and more stable driving behavior. This framework contributes to the development of transparent and trustworthy autonomous driving systems suitable for safety-critical applications

This research investigates integrated maritime workforce resilience and mental health management frameworks addressing post-pandemic seafarer wellbeing challenges and organizational safety culture transformation. Through qualitative analysis involving 39 stakeholders including seafarers, ship operators, mental health professionals, maritime unions, training institutions, and maritime authorities, this study examines how COVID-19 pandemic intensified mental health crises through extended contracts, shore leave restrictions, and isolation while exposing systemic inadequacies in psychological support systems. Results demonstrate that comprehensive mental health frameworks can reduce psychological distress by 55-70%, improve safety performance by 40-55%, enhance crew retention by 45-60%, and decrease incident rates by 35-50% when integrating organizational culture change, leadership competency development, predictive analytics, and culturally-adapted interventions. Key challenges include mental health stigma (affecting 65-80% of seafarers), limited organizational investment (only 18-25% adequate), service accessibility gaps, and workforce demographic diversity requiring culturally-sensitive approaches. Findings reveal that effective mental health management requires systemic organizational transformation integrating psychological wellbeing into safety management systems, work design optimization, family support programs, and career sustainability rather than treating mental health as peripheral welfare concern, supporting maritime industry's workforce retention and operational safety imperatives.

This research proposes an alternative model for predicting the plasticity potential of clay soils, updating the previous model developed by Firincioglu and H. Bilsel. The alternative lies in the use of the fine fraction content (FC) as the predictive variable, replacing the percentage of sand fraction content (SC) previously suggested. The analysis was conducted on 61 low-plasticity clay (CL) soil samples, classified using the Casagrande and Moreno-Maroto systems, by examining the relationship between consistency limits (LL, PI), grain fractions (sand, silt, clay), and related parameters (FC, sand-silt-clay spectrum, and sand fraction ratio). The correlation analysis results show significant findings, including a strong positive correlation between sand content and the sand ratio (SR), as well as a negative correlation between sand content and FC, and between FC and SR. The performance of the substitute Casagrande quadratic model [2] reveals the best predictive accuracy among the proposed models ( [1], R² =0.90; [3], R²  0.89; [4], R² =0.43; [5], R² =0.93; [6], R² =0.93; [7], R² =0.89; [8], R² =0.89), with the highest R2 value of 0.93, MSE of ≈4.01, and MAE of ≈1.44–1.47. The equation is .

Accurate wind turbine power curve modeling plays a crucial role in performance evaluation, energy yield estimation, and data-driven control strategies. However, actual power curves often exhibit non-linear behavior influenced by atmospheric variability, measurement noise, and SCADA anomalies, making conventional modeling approaches less effective. This study proposes an optimized logistic power curve model whose parameters are tuned using Particle Swarm Optimization (PSO) to improve predictive accuracy. The analysis uses the Wind Turbine SCADA Dataset from Kaggle, which undergoes extensive preprocessing including physical rule filtering, outlier detection with the Interquartile Range (IQR) method, anomaly removal, and smoothing of the power signal. A three-parameter logistic model is selected due to its ability to capture the typical S-shaped relationship between wind speed and power output. PSO is applied to identify optimal model parameters by minimizing the Mean Squared Error (MSE), utilizing 40 particles over 200 iterations. The optimized model achieves strong predictive performance with RMSE of 404.09, MAE of 179.96, and R² of 0.904 on the test set, indicating that more than 90% of the variability in actual power can be explained by wind speed. Residual analysis reveals heteroscedastic patterns and slight overestimation in mid-range wind speeds, yet overall model consistency remains high. Comparative evaluation against Linear Regression, Random Forest, and logistic modeling using curve_fit shows that the Logistic–PSO approach provides the most accurate and stable predictions. These findings demonstrate that combining logistic modeling with PSO offers an effective and robust method for data-driven wind turbine power curve optimization.

Chronic kidney disease is an increasingly prevalent health issue that requires more precise clinical data-based early detection methods to enable timely and appropriate treatment. This study focuses on developing a predictive model for chronic kidney disease using the Light Gradient Boosting Machine (LightGBM) algorithm and enhancing its performance through hyperparameter optimization with the Grey Wolf Optimizer (GWO). The dataset used originates from public sources and undergoes several preprocessing steps, including missing value imputation, categorical feature encoding, outlier handling, initial feature selection, and stratified data splitting to maintain model quality. Three modeling approaches were evaluated: LightGBM with default parameters, LightGBM enhanced using Random Search, and LightGBM optimized with GWO. The experimental results indicate that the baseline model already performs well, Random Search improves accuracy and F1-score, and GWO achieves the highest AUC-ROC value despite requiring longer computation time. Significance testing through cross-validation shows that the performance differences among the three models are not statistically significant, suggesting that the observed improvements are not strong enough to determine a definitively superior optimization method. The feature importance analysis highlights that clinical indicators such as creatinine levels, glomerular filtration rate, blood pressure, and urine protein contribute most prominently to the prediction. Overall, the study demonstrates that LightGBM is a reliable model for early detection of chronic kidney disease, and hyperparameter optimization still offers added value that can support the development of AI-based clinical decision-support systems

This study analyzes the implementation of ESG (Environmental, Social, and Governance) reporting on the firm value of PT Indofood Sukses Makmur Tbk during the 2020-2024 period. The main issues examined are how ESG reporting is implemented and the extent of its influence on firm value, as well as which ESG component has the most significant impact. The research method employs a quantitative approach with a causal comparative design, utilizing secondary data sourced from annual reports, sustainability reports, and market data over five years. Firm value is proxied using Tobin's Q ratio, while the level of ESG disclosure is measured based on the GRI Standards framework. Data analysis techniques use multiple linear regression by incorporating control variables including firm size, profitability (ROA), and leverage to enhance result validity. The research findings indicate that ESG reporting has a positive and significant effect on firm value with a coefficient of β = 0.018 and p < 0.001, with a model predictive capability (R²) of 87.3%. Indofood's ESG Score experienced substantial improvement from 56.3% in 2020 to 78.9% in 2024, accompanied by an increase in Tobin's Q from 0.982 to 1.523. Component-wise analysis reveals that the Social aspect provides the highest impact (β = 0.009), followed by Governance (β = 0.007) and Environmental (β = 0.006). These findings provide empirical support for stakeholder theory and resource-based view in the Indonesian emerging market context.

This study aims to examine and analyze the influence of debt maturity and cash holding on dividend policy, the effect of debt maturity and cash holdings on company value, and the effect of debt maturity and cash holdings on company value through dividend policies in banking sub-sector companies listed on the Indonesia Stock Exchange for the 2020-2023 period. The sampling technique uses the purposive sampling method. Secondary data was obtained from the annual financial statements. The data analysis method uses path analysis with two structural equations to test the direct and indirect influence of independent variables on dependent variables through intervening variables. The results of the structural equation 1 study show that debt maturity has no effect on dividend policy, and cash holdings have an effect on dividend policy. The results of structural equation 2 show that debt maturity affects the value of the company and cash holdings have no effect on the value of the company. The results of the testing of intervening variables show that dividend policy can mediate debt maturity to company value and dividend policy can mediate cash holdings to company value. The predictive ability of the two variables on the dividend policy was 13.6% and the remaining 12.6% was influenced by other variables outside the research model and the predictability of the three variables on the company's value was 15.7% and the remaining 14.7% was influenced by other variables outside the research model.

This study examines the relationship between household literacy, access to basic infrastructure, and the quality of primary education in Eastern Indonesia within the framework of Christian Religious Education values. Using a quantitative explanatory design, the research analyzes national socio-economic data with provinces as the unit of analysis. The model was tested using Partial Least Squares Structural Equation Modeling (PLS-SEM) to evaluate the predictive power of household access to safe water, access to electricity, and literacy level on primary education outcomes, measured by the average national exam score. The results show that access to safe water (β = 0.294, p < 0.05) and electricity (β = 0.290, p < 0.05) have significant positive effects on education quality, while literacy level (β = 0.081, p > 0.05) is not statistically significant. The model explains 34.7% of the variance in education quality (R² = 0.347). These findings indicate that improving basic infrastructure remains essential to enhancing educational outcomes in Eastern Indonesia. Integrating Christian educational values such as justice, love, and service can further strengthen community motivation and collective responsibility toward equitable education development.

The Human Development Index (HDI) functions as a key indicator for assessing the level of welfare and overall quality of life of the population within a specific region. This study aims to examine the socio-economic factors influencing HDI at the provincial level in Indonesia using a Gaussian kernel regression approach. A nonparametric method is employed due to its flexibility in capturing nonlinear relationships between the response and predictor variables without the need to assume a specific functional form. The analysis utilizes secondary data, including education, poverty, per capita expenditure, expected years of schooling, open unemployment rate, and gross regional domestic product for each Indonesian province. The findings from this study indicate that educational factors, particularly mean years of schooling and expected years of schooling, exert the most significant impact on HDI improvement. The estimated Gaussian kernel regression model demonstrates a coefficient of determination of 0.9954 and a residual standard error of 0.3468, reflecting a very high predictive accuracy and relatively low error. These results suggest that Gaussian kernel regression is an effective nonparametric approach for analyzing human development in Indonesia.

This study examines the effectiveness of the Early Warning System (EWS) in anticipating and responding to mental health crises in conflict-affected regions of the Middle East through a systematic review of 47 scholarly articles published between 2014 and 2024. The meta-regression findings indicate a significant contribution of EWS implementation to the reduction of post-traumatic stress disorder (PTSD) symptoms with a coefficient of β = -0.67 (p < .001), as well as depressive symptoms with a coefficient of β = -0.59 (p < .001) among populations directly affected by armed conflict. Among 12,456 respondents analysed, 73.8% reported a reduction in anxiety symptoms following the implementation of EWS, with an effect size of d = 0.82 (95% CI [0.76, 0.88]). Digitally based early warning systems demonstrated a significantly higher level of effectiveness (OR = 2.34, 95% CI [1.98, 2.70]) than conventional systems, which are more manual and reactive. Moderator analysis indicated that age (β = -0.31, p < .01) and the duration of exposure to conflict (β = 0.44, p < .001) play important roles in moderating the relationship between EWS interventions and various mental health indicators. These findings expand upon the conclusions of Fu et al. (2020) and Salesi (2023), which previously explored psychosocial interventions in conflict zones, by adding a new dimension—examining digital technology and predictive algorithms within EWS frameworks. The study explicitly demonstrates that integrating machine learning models into EWS can enhance the predictive accuracy of potential mental health crises to 84.6%, representing a novel contribution that has not been comprehensively documented in prior academic literature

This study aims to analyze the influence of academic achievement, internship experience, competence, and work motivation on the work readiness of Gunadarma University Management students in the Depok region. The research employed a quantitative method using primary data collected through questionnaires. A total of 250 valid responses were obtained using a non-probability sampling technique with purposive sampling. Data analysis was conducted using Smart PLS 4 software with several testing stages, including convergent validity, Average Variance Extracted (AVE), composite reliability, Cronbach’s alpha, discriminant validity, R-square test, predictive relevance (Q²), and hypothesis testing. The results indicate that academic achievement, internship experience, competence, and work motivation significantly influence students’ work readiness, both directly and indirectly. These findings highlight the importance of enhancing practical experience and internal motivation among students to better prepare them for the increasingly competitive job market.

In the ever-evolving business landscape, student work readiness plays a vital role as a foundation for successful entry into the workforce. This study aims to examine the factors influencing student work readiness by analyzing the effects of academic achievement, internship experience, competence, and work motivation among management students at Gunadarma University Kalimalang. The research employs a quantitative approach using primary data collected through questionnaires distributed to 250 respondents. Data were analyzed using SmartPLS 4.0 with various statistical tests, including convergent validity, discriminant validity, Average Variance Extracted (AVE), R-Square, predictive relevance (Q²), and path coefficient analysis. The findings reveal that academic achievement and work motivation significantly and positively affect students’ work readiness, indicating that strong academic performance and high motivation enhance preparedness for entering the workforce. Conversely, internship experience and competence do not have a significant impact on students’ work readiness, suggesting that practical experience and skill development may not be fully integrated into the academic process. This study underscores the importance of fostering motivation and maintaining academic excellence to improve graduate employability and readiness in a competitive job market.

This research was carried out with the primary objective of developing a standardized anthropometric testing model to significantly improve the performance of divers in Pasuruan Regency. This developed model serves as an essential predictive and evaluative tool in the context of athlete training and selection. The study employed a descriptive quantitative method, involving a total of 17 active divers, comprising 10 male athletes and 7 female athletes. Data collection was conducted through a series of basic anthropometric measurements relevant to diving sports, including body height, body weight, arm span, and leg length. Comprehensive statistical analysis revealed a significant and positive relationship between the various measured anthropometric variables and key diver performance parameters. The main findings of the research clearly indicate that the developed anthropometric testing model has a strong and significant correlation with the enhancement of the divers' strength, speed, and endurance. In other words, this model is capable of providing accurate insights into the potential and actual performance status of the divers. The practical implications of this study are highly significant, as it enriches knowledge and technological innovation in the field of diving sports training. Furthermore, the model offers concrete support for coaches in more objective athlete selection, the design of more personalized and structured training programs, and indirectly contributes to the improvement of a diver's quality and self-confidence. Therefore, this anthropometric testing model is recommended for widespread use as a strategic tool in all stages of selection, training, and long-term development of diving athletes.

This study aims to compare and analyze the accuracy levels of four financial distress prediction models—Altman Z-Score, Springate, Grover, and Zmijewski—in anticipating the potential bankruptcy of companies subjected to delisting from the Indonesian Stock Exchange (IDX). The delisting phenomenon, which is strongly linked to severe financial deterioration, provided the core motivation for identifying the most reliable predictive instrument, utilizing secondary data from the annual financial reports of delisted companies during the 2019-2023 observation period. Descriptive analysis techniques were employed to calculate the accuracy rate and Type Error for each model. The comparative results consistently indicate that the Springate Model is the most effective, consistent, and accurate model for predicting financial distress in delisted firms, achieving an accuracy rate of 89% in both the first and second years prior to delisting, while the Altman Z-Score model exhibited lower accuracy (68.75% and 62.50%). This key finding emphasizes the superiority of the Springate Model as a crucial diagnostic tool for investors and regulatory bodies in assessing corporate bankruptcy risk.