Publication Search

This study aims to compare the performance of three boosting algorithms, namely XGBoost, LightGBM, and CatBoost, to predict the success of MSMEs. The data used consists of 250 entries with 13 attributes that include business actor characteristics, initial capital, industry experience, financial record-keeping, internet utilization, business planning, partnerships, and the target variable success. The pre-processing stage includes checking for missing values, standardizing numerical attributes, and splitting the data into 80% training data and 20% test data. The evaluation results show that XGBoost provides the best performance with an accuracy of 0.92, precision of 0.8333, recall of 0.8333, F1-score of 0.8333, and ROC-AUC of 0.9715. LightGBM has an accuracy of 0.88, while CatBoost achieves an accuracy of 0.90. The research results show that XGBoost has the best ability to classify successful and unsuccessful MSMEs. The feature importance results also show that the success of MSMEs is influenced by a combination of several key factors. This research emphasizes that boosting algorithms are effectively used as predictive models to support the analysis of MSME success.

This study investigates the integration of digital twin technology and machine learning for predictive analysis in smart mechanical systems. The research emphasizes the role of intelligent computational frameworks in improving industrial monitoring, predictive maintenance, and operational efficiency within Industry 4.0 environments. A qualitative content analysis approach was employed by reviewing scientific literature, industrial reports, and previous studies related to digital twins, artificial intelligence, and predictive analytics. The findings indicate that digital twin architectures supported by machine learning algorithms can significantly enhance real-time monitoring, fault prediction accuracy, and maintenance optimization. The integration of IoT devices, cloud computing, and intelligent analytics also improves industrial sustainability, reduces operational downtime, and supports data-driven decision-making processes. Furthermore, the study identifies several technological challenges, including cybersecurity risks, data integration complexity, and computational limitations. Overall, the proposed intelligent digital twin framework provides a promising approach for future industrial innovation and sustainable smart mechanical system management

Hypertension is one of the leading non-communicable diseases contributing significantly to cardiovascular morbidity and mortality worldwide. Despite the availability of extensive electronic medical record data in healthcare institutions, these data are often utilized only for administrative reporting rather than predictive analysis. Consequently, opportunities to identify age groups with a higher probability of developing hypertension remain underutilized. This study aims to implement the Naïve Bayes classification algorithm to analyze age distribution and classify the risk of hypertension among patients using healthcare data. The research adopted the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology, including business understanding, data understanding, data preparation, modeling, evaluation, and deployment. Patient medical record data consisting of demographic and clinical attributes, including age, systolic blood pressure, diastolic blood pressure, body weight, gender, and hypertension status, were processed using the Naïve Bayes algorithm. Model performance was evaluated using a confusion matrix by measuring accuracy, precision, recall, specificity, and balanced accuracy. The implementation demonstrates that the Naïve Bayes algorithm is capable of classifying hypertension risk efficiently while providing probabilistic information regarding age groups with a higher tendency to experience hypertension. The resulting classification model offers an effective decision-support tool for healthcare providers in conducting targeted screening, preventive interventions, and evidence-based health planning. The findings also indicate that data mining techniques can transform routinely collected medical records into valuable clinical knowledge for early hypertension prevention and healthcare decision-making.

This study explores the role of artificial intelligence in strengthening cybersecurity threat detection frameworks for next-generation network environments. The rapid expansion of cloud computing, Internet of Things ecosystems, and distributed digital infrastructures has significantly increased cybersecurity risks and operational vulnerabilities. Traditional cybersecurity systems often struggle to detect sophisticated and evolving threats due to their dependence on static detection mechanisms. Using a qualitative research approach and content analysis method, this study examines recent developments in artificial intelligence, machine learning algorithms, and intelligent cybersecurity frameworks. The findings indicate that AI-driven cybersecurity systems improve real-time threat detection, anomaly identification, automated monitoring, and predictive security analysis. Machine learning technologies such as Random Forest, Support Vector Machine, and deep learning models demonstrate strong potential for enhancing intrusion detection accuracy and reducing false positive rates. The study also identifies critical challenges related to ethical governance, privacy protection, computational complexity, and adversarial attacks in AI-based cybersecurity systems

Stunting is a multifactorial public health problem influenced by various risk factors that may emerge during the prenatal period. Early identification of stunting risk during pregnancy is important to support preventive interventions. This study aimed to develop a stunting risk prediction model based on maternal prenatal factors using the Random Forest algorithm. Secondary data from 172 pregnant women, consisting of 83 stunting cases and 89 non-stunting cases, were analyzed. The predictor variables included maternal age during pregnancy, height, hemoglobin level, mid-upper arm circumference (MUAC), smoking history, hypertension, asthma, and diabetes mellitus. The research stages consisted of data preprocessing, model training using Stratified 5-Fold Cross Validation, performance evaluation, external testing, and feature importance analysis. Internal evaluation results showed an accuracy of 60%, precision of 60.6%, recall of 57.3%, F1-score of 58.9%, and AUC of 0.6688. External testing yielded an accuracy of 70% and an AUC of 0.6167. Feature importance analysis identified maternal age during pregnancy as the most influential variable in the prediction process. The findings indicate that maternal prenatal factors have potential for early stunting risk identification, although the predictive performance remains moderate. This approach may serve as a foundation for developing early screening tools to support targeted interventions among high-risk pregnancies.

The phenomenon of tourist destinations experiencing a decline in popularity after a viral phase (post-viral stagnation) poses significant challenges to the sustainability of urban tourism. This study aims to test a structural model for destination reactivation in Kampung Pelangi, Semarang, focusing on the role of digital transformation in strengthening destination resilience through the mediation of competitive advantage. Using a quantitative approach, data were collected from 150 respondents and analyzed using Structural Equation Modeling (SEM-PLS). The results of the hypothesis testing indicate that digital transformation has a positive and significant influence on competitive advantage (β = 0.495; t = 5.820; p < 0.001) and destination resilience (β = 0.312; t = 3.450; p < 0.001). Furthermore, competitive advantage was found to have a strong impact on resilience (β = 0.542; t = 7.115; p < 0.001). A mediation test demonstrated that competitive advantage significantly mediated the relationship between digital transformation and resilience (β = 0.268; t = 4.890; p < 0.001). This model demonstrated a predictive power of 61.2% (R2 = 0.612) for destination resilience. This finding emphasizes that post-pandemic destination reactivation requires the integration of digital capabilities that can restore differentiation and unique value propositions to achieve long-term resilience.

.This study aims to analyse the role of Financial Information Systems (FIS) in supporting risk management, decision-making, and organisational performance in the digital transformation era. This study employs the Systematic Literature Review (SLR) method to examine articles indexed in Scopus from 2016 to 2026. The PRISMA framework is used to ensure a systematic, transparent article selection process, resulting in the selection of 37 relevant articles for further analysis. The results of the study show that Financial Information Systems make a major contribution to improving financial transparency, operational efficiency, the quality of strategic decision-making, and organisational risk mitigation. In addition, the integration of emerging technologies such as Artificial Intelligence (AI), FinTech, big data analytics, and cloud computing further strengthens the effectiveness of financial information systems in modern organisations. This study contributes theoretically by mapping research trends and identifying research gaps, while providing practical benefits for organisations seeking to increase competitiveness through digital financial systems. For future research, it is recommended to develop a more predictive and intelligent Financial Information Systems model to address future business dynamics.

Artificial intelligence has the potential to support radiology workflows by assisting in the identification of cases that may require additional clinical attention. However, alert-oriented medical AI systems should provide not only classification outputs but also interpretable evidence that can be reviewed and audited by clinicians. This study develops and evaluates an explainable multimodal framework for binary chest X-ray alert classification using paired radiology reports and chest X-ray images. The text branch employs TF-IDF n-gram features with a class-balanced Logistic Regression classifier, while the image branch fine-tunes a pretrained ResNet18 model. The two branches are integrated through probability-level late fusion using a validation-selected fusion weight. Explainability is implemented in a modality-specific manner: global coefficient analysis is used to identify influential textual cues, while Grad-CAM heatmaps are used to visualize salient image regions. Experiments were conducted on paired samples from the Open-i/IU X-Ray dataset using text-only, image-only, and fusion-based evaluation settings. Additional analyses include case-level complementarity analysis, bootstrap confidence intervals for ROC-AUC, shortcut-feature inspection, and qualitative Grad-CAM auditing. The results indicate that the text modality provides the dominant predictive signal under the current proxy-label setting. Late fusion produced a small descriptive improvement on the test set, increasing accuracy from 0.8533 to 0.8667, F1-score from 0.8817 to 0.8936, and ROC-AUC from 0.8936 to 0.9025 compared with the text-only baseline. However, the observed ROC-AUC improvement was not statistically conclusive based on bootstrap analysis. These findings suggest that the proposed framework is useful as a reproducible and auditable multimodal prototype, while also highlighting important limitations, including proxy-label ambiguity, potential label leakage from radiology reports, limited image-branch contribution, lack of external validation, and the need for stronger explanation and calibration assessment.

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 advancement of digital technology has significantly transformed organizational decision-making, particularly in modern leadership contexts that demand rapid and data-driven responses. Artificial Intelligence(AI) has emerged as a strategic technology capable of enhancing accuracy, speed, and effectiveness in decision-making through comprehensive data analysis. This study aims to analyze the role of AI in supporting leadership decision-making and its implications for organizational effectiveness using a narrative literature review approach. Secondary data comprising peer-reviewed national and international journal articles were analyzed to identify patterns, themes, and interactions between AI, leadership, and decision-making processes. The findings indicate that AI functions not only as a data analysis tool but also as a strategic element that strengthens leaders’ capabilities in evidence-based decision-making, improves team coordination, and optimizes organizational processes. Thematic synthesis identified three main domains analytics and predictive capabilities, leadership strategies, and implementation challenges that form the basis for integrating AI into managerial practice. This study contributes theoretically by expanding the digital leadership and technology-based decision-making framework and practically by providing guidance for organizations to optimize AI utilization to enhance decision quality and efficiency. The research also offers directions for future empirical studies to explore AI-leadership interactions across various organizational sectors, supporting more adaptive, effective, and data-driven decision-making in the digital era.

This study analyzes the performance of three types of controllers—PID, Fuzzy-PID, and Model Predictive Control (MPC)—in maintaining water level stability in a dam system. Proper water level regulation is essential to prevent flooding, ensure adequate water distribution, and maintain the structural safety of the dam. Previous studies have widely implemented PID and Fuzzy-PID controllers; however, MPC offers significant advantages through system prediction and multivariable optimization. This research compares the performance of the three controllers through simulation analysis using a second-order mathematical model. The results indicate that MPC has the potential to deliver a faster settling time and a smaller steady-state error compared to PID and Fuzzy-PID, although it requires higher computational complexity.

Individual behavioral factors are critical determinants of organizational effectiveness and a vital component of modern organizational diagnosis. This study aims to identify and map individual behavioral factors as strategic indicators in organizational diagnosis using a Systematic Literature Review (SLR) approach guided by PRISMA 2020. Analysis of ten selected articles reveals that organizational effectiveness in the digital transformation era is driven by a reciprocal equilibrium between an individual's cognitive infrastructure and volatile work environment demands. The findings synthesize these behaviors into four strategic clusters: psychological well-being as primary infrastructure, digital structural support audits, justice and trust equilibrium, and psychological contract synchronization. This research provides tactical implications for Human Capital practitioners to transform annual diagnostic methodologies toward the implementation of monthly pulse surveys to detect fluctuations in well-being and disengagement intentions in real-time. This predictive diagnostic step is essential to mitigate turnover risks and design precise institutional interventions in hybrid work ecosystems.

This study examines customer churn prediction in subscription-based telecommunications from a digital marketing perspective using machine learning. The analysis utilizes a secondary dataset of 7,043 customer records that simulate behavioral, contractual, and financial attributes commonly found in telecom services. Three classification algorithms Logistic Regression, Random Forest, and Gradient Boosting are applied to model churn behavior. Data preprocessing includes handling missing values, encoding categorical variables, and splitting data into training and testing sets. Model performance is evaluated using accuracy, recall, and ROC-AUC, with emphasis on recall due to its importance in identifying at-risk customers. The results show that Gradient Boosting achieves the highest overall performance with an ROC-AUC of 0.84, while Logistic Regression provides relatively higher recall. Key drivers of churn include short-term contracts, higher monthly charges, and lower service engagement. However, recall remains moderate, indicating limitations in capturing complex behavioral factors. These findings suggest the need to combine predictive models with behavioral insights and highlight the importance of early customer engagement and long-term retention strategies.

Edentulism is a condition of total tooth loss that has a significant impact on the efficiency of mastics, phonetics, and facial aesthetics. This case report presents the rehabilitation of a 72-year-old female patient with a condition of total mandible edentulus using two standard-diameter implants that support the overdenture with a locator retention system. Treatment procedures include clinical evaluation, radiographic analysis, implant placement, prosthesis placement, as well as follow-up evaluation to assess function and comfort. The results of the treatment showed an increase in patient retention, stability, and comfort in daily activities. In addition, patients reported improved confidence and quality of life after the use of implant-based overdentures. These findings confirm that overdenture with mandibular implant support is a predictive, effective, and reliable rehabilitation method in treating total edentulism. The success of this case provides clinical evidence that implant-based approaches are able to overcome the limitations of conventional prostheses, as well as being a solution that supports the functional and psychosocial aspects of elderly patients. Thus, implant-based overdenture can be recommended as the primary therapeutic option in the rehabilitation of mandibular edentulism.

This study aims to analyze the partial and simultaneous effects of Investment Decisions, Financing Decisions, and Company Size on Company Value at PT Ciputra Development Tbk for the period 2014-2024. Company value is proxied by Price to Book Value (PBV), investment decisions by Price Earning Ratio (PER), financing decisions by Debt to Equity Ratio (DER), and company size by SIZE. The method used in this study is descriptive quantitative. The population of this study is the financial statements of PT Ciputra Development Tbk for the period 2014-2024, and the sample used is the financial position report, income statement, and share price of PT Ciputra Development Tbk for the period 2014-2024. The analysis methods used are descriptive analysis, classical assumption testing, multiple linear analysis, t-test, f-test, and coefficient of determination test using SPSS version 26. The results of the study show that partially, PER and DER do not have a significant effect on PBV, while SIZE has a negative and significant effect on PBV. Simultaneously, PER, DER, and SIZE significantly affect PBV with a coefficient of determination of 94.7%, indicating that the regression model has excellent predictive power. The remaining 5.3% is influenced by other variables outside the scope of this study.

Indonesia harbors extraordinary linguistic diversity with over 700 regional languages representing approximately 10% of the world's languages within 1.3% of global land area. However, this diversity faces existential threat from language shift toward Indonesian, urbanization, education policies favoring the national language, and globalization. UNESCO classifies 146 Indonesian languages as endangered, with several dozen facing imminent extinction as last speakers age without intergenerational transmission. This study documents the current vitality status of Indonesian regional languages, analyzes factors driving language endangerment and shift, evaluates existing conservation efforts, and proposes evidence-based strategies for language revitalization and maintenance. A multi-phase approach was employed: vitality assessment of 150 regional languages using UNESCO's Language Vitality and Endangerment framework with surveys involving 2,400 speakers; ethnographic case studies in 12 speech communities; policy analysis; evaluation of 25 revitalization programs; and predictive modeling of language shift trajectories. Of 150 surveyed languages, only 23 (15.3%) classified as safe with robust intergenerational transmission; 48 (32.0%) were vulnerable; 42 (28.0%) definitely endangered; 28 (18.7%) severely endangered; and 9 (6.0%) critically endangered. Key endangerment drivers included Indonesian-only education (92.3% of schools), urban migration (67.8% of youth), negative language attitudes (54.2% of parents), and lack of written traditions (73.4% of languages lacking orthographies). Modeling projected that without intervention, 40% of currently vulnerable languages will become definitely endangered within 20 years. Successful revitalization demands community-owned interventions, mother-tongue-based multilingual education, new digital language domains, and attitude change campaigns. Indonesia's linguistic diversity represents invaluable cultural and scientific heritage requiring urgent, coordinated conservation action.

High-dimensional clinical data exhibit complex and non-linear relationships among patient attributes, where outcomes are often influenced by feature interactions rather than isolated variables. However, many existing machine learning models prioritize predictive performance while providing limited interpretability and insufficient insight into interaction structures. This study aims to address this limitation by developing an interpretable and robust framework for feature interaction mining in clinical data. We propose a hybrid tree–neural modeling framework that explicitly captures and ranks feature interactions while maintaining stable predictive performance. Tree-based ensemble models are employed to identify non-linear interaction patterns, while neural representations enhance learning flexibility and generalization. The framework integrates interaction importance analysis, cross-validation–based stability assessment, and evaluation across multiple data splits to ensure robustness and interpretability. Experiments conducted on a real-world high-dimensional clinical dataset demonstrate that the proposed approach achieves consistent predictive performance, with AUC values ranging from 0.628 to 0.641 across five cross-validation folds (mean AUC ≈ 0.633). Performance remains stable under varying train–test splits, indicating strong generalizability. Interaction analysis reveals that a small number of dominant feature interactions—such as age combined with length of hospital stay and medication count combined with diagnostic information—consistently contribute to model predictions, appearing in over 80% of validation folds. Ablation studies further confirm that removing interaction-aware components leads to noticeable performance degradation, highlighting their importance.  In conclusion, this study demonstrates that explicit feature interaction modeling enhances interpretability, stability, and generalization in clinical prediction tasks. The proposed hybrid framework provides a reliable foundation for developing trustworthy and transparent clinical decision-support systems

This meta-analytic study aims to comprehensively examine the relationship between mental health, learning capacity among health education students, and competencies in nosocomial disease risk management through cross-contextual empirical synthesis. An analysis of 47 studies involving 12,847 participants from 15 countries demonstrates a strong, statistically significant association between students' mental health and competencies in nosocomial infection prevention, as reflected by a correlation coefficient of r=0.68 (p<0.001) and a 95% confidence interval of 0.61-0.74. Students with high mental health scores (M=78.4; SD=8.2) exhibited substantially superior understanding of infection prevention protocols, namely 43% higher than the control group (M=54.7; SD=12.1; t(846)=18.42; p<0.001; d=2.31). Structural equation modeling confirmed learning capacity as a significant partial mediator (β=0.52; p<0.001), with an indirect effect reaching 35.4% and a 95% CI range of 28.6-42.1%. Mindfulness-based psychoeducational interventions were shown to enhance nosocomial risk identification abilities by 38.7% (F(2,564)=42.18; p<0.001; η²=0.41) while reducing clinical anxiety by 31.2% (t(382)=9.84; p<0.001). These findings extend the frameworks proposed by Song (2024) and Schutte et al. (2025), which primarily emphasize cognitive aspects, by demonstrating that the integration of psychological dimensions yields a multidimensional predictive model explaining 64.3% of the variance in risk management competence (R²=0.643; F(5,841)=304.76; p<0.001), surpassing conventional models that account for only 38-45% of the variance.

This study aims to provide a comprehensive analysis of Predictive Policing and its implications for law enforcement transformation in Indonesia, based on an extensive review of its global applications, benefits, and challenges. The study uses qualitative literature and international case study review methods to assess the impact and complexity of implementing digital technologies such as artificial intelligence (AI), machine learning, and big data analytics within a Predictive Policing framework. The results of this review highlight that while Predictive Policing offers significant potential for proactive crime prevention and increased operational efficiency, its implementation is consistently fraught with critical legal, ethical, and technical challenges, including regulatory gaps, risks of algorithmic bias, and data privacy concerns, which are particularly relevant to Indonesia. The findings underscore that public trust and police legitimacy in the context of adopting such technologies are strongly influenced by transparency, strong accountability mechanisms, and community involvement in shaping their use. This study contributes to the growing discourse on digital policing in developing countries and culminates in practical policy recommendations designed to guide the Indonesian police towards the development and implementation of Predictive Policing models that are effective, efficient, and fundamentally respectful of legal and human rights principles.

Pneumonia remains a leading cause of morbidity and mortality worldwide, particularly in resource-limited settings and among elderly populations, where timely diagnosis and continuous monitoring are often constrained by limited clinical infrastructure. This study presents an edge–cloud–integrated framework for early pneumonia risk monitoring, leveraging multimodal wearable sensors and deep learning to support continuous short-duration monitoring. The proposed system is designed to operate in near real time under simulated deployment conditions, continuously acquiring and analyzing physiological signals (respiratory rate, heart rate, SpO₂, and body temperature) alongside event-driven acoustic biomarkers (cough sounds) within a distributed architecture. A lightweight edge module performs local signal preprocessing and anomaly triage, selectively transmitting salient information to a cloud-based multimodal deep learning model for refined risk estimation and interpretability analysis. The framework was evaluated using a multi-source dataset comprising public repositories (MIMIC-III and Coswara) and a clinically supervised wearable study conducted in two Nigerian hospitals, resulting in 718  hours of quality-controlled multimodal monitoring data. In a pooled multi-source evaluation, the system achieved an AUC of 0.95, while in a clinically realistic local-only evaluation, the AUC was 0.86, reflecting a consistent but preliminary diagnostic signal. These results highlight the importance of local data adaptation for real-world applicability and suggest that multimodal AI can provide meaningful early risk indicators under resource constraints. Beyond predictive performance, this work demonstrates the feasibility of integrating multimodal learning, edge–cloud computation, and explainable analytics into a deployment-aware, privacy-preserving monitoring framework for low-resource healthcare environments.