To solve the problem of imbalanced resource allocation in universities, we take six secondary colleges of a comprehensive university as the research object and propose a university resource optimization and governance decision-making method based on multidimensional data fusion. For data fusion, the data layer completes the cleaning and unified mapping of heterogeneous data from multiple sources, and the feature layer performs quality evaluation on the integrity rate and cross system consistency verification of the introduced data. Objectively determine the weights of each indicator using the entropy weight method. When constructing the indicator system, it covers three dimensions: supply, efficiency and fairness, and verifies the internal consistency and structural validity of the indicator system through Cronbach’s alpha and KMO tests. For resource optimization, we formalize the problem of cross departmental teacher allocation as a constraint optimization model, with the objective function of minimizing the coefficient of variation of the student teacher ratio. We set three types of constraints and directly use the model output as the basis for allocation decisions. Scenario comparison analysis shows that after optimization, the coefficient of variation of teacher to teacher ratio decreased by 40.2%, classroom utilization rate increased by 9.76%, and the comprehensive index of each college exceeded the warning threshold of 0.40.
Persistent challenges in teachers’ digital readiness continue to highlight the need for practical digital skills required for effective instructional practice. This study aimed to (1) develop a microlearning-integrated blended learning model delivered via an online platform, (2) evaluate its instructional efficiency, and (3) examine its effectiveness in enhancing pre-service teachers’ digital skills. A research and development approach was employed using a one-group pretest–posttest design with 268 pre-service teachers over six weeks. Instruction was implemented in a 50:50 online–offline format. Data were analyzed using descriptive statistics and paired t-tests to assess changes in digital skills across six performance-based domains. The results showed that the developed MB Learning Model achieved an instructional efficiency of E1/E2 = 90.88/86.76, exceeding the established 80/80 benchmark. Significant improvements were observed across all six digital skill domains (p < 0.001), with the overall digital skill score increasing from M = 39.52 to M = 54.53. These findings indicate substantial gains in both declarative knowledge and performance-based competencies. The findings suggest that the structured integration of microlearning within a blended learning environment delivered via an online platform is associated with improvements in digital skill development. However, due to the absence of a control group, the results should be interpreted with caution, as causal relationships cannot be definitively established. Despite this limitation, the model demonstrates practical potential as a structured instructional framework for supporting digital skill development in teacher education contexts.
The kompang is a traditional single-headed frame drum that holds significant cultural value within the Malay Archipelago. However, modernization and the proliferation of digital entertainment technologies have contributed to a decline in both its popularity and the number of skilled practitioners. Contributing factors include limited access to qualified instructors, fragmented scheduling and coordination of training sessions, and the absence of interactive learning platforms. This study presents Pak Bung, an intelligent chatbot system designed to promote cultural preservation and public engagement with kompang music through adaptive digital learning. The chatbot Leverages Large Language Models (LLMs), cloud-based knowledge retrieval, and interactive dialogue systems to deliver personalized educational content across seven curated topics. Features include multimodal tutorials, gamified quizzes, multimedia demonstrations, and real-time scheduling for consultations with human experts. A chatbot usability testing was conducted to assess usability, engagement, and perceived usefulness for learning, with results indicating that Pak Bung can successfully support user engagement, accessibility, and perceived learning support in cultural knowledge exploration. The system aligns with Sustainable Development Goal 4 (Quality Education) by integrating AI-driven pedagogical interaction into the preservation of traditional musical knowledge.
The shift to flexible learning during the Philippine natural calamities prompted higher education institutions to adopt alternative instructional modalities, particularly modular and online learning. This study examined the students’ evaluations of these modalities at the College of Teacher Education at the Abra State Institute of Sciences and Technology. It assessed students’ perceptions and determined the significant differences across academic programs. A descriptive-comparative research design was employed involving the 179 third-year students. Data were collected using a researcher-developed questionnaire validated through the Content Validity Index (CVI) with an average of 0.89 and demonstrating acceptable reliability (Cronbach’s α = 0.74). Descriptive statistics, including mean and standard deviation, were used to describe students’ evaluations. At the same time, Analysis of Variance (ANOVA) was applied to determine significant differences among programs. Findings revealed that students generally had positive evaluations of both modular and online learning. However, issues related to internet connectivity and access to learning resources were identified as major challenges in online learning. Statistical analysis showed a significant difference in students’ evaluations of modular learning across programs, while no significant difference was found in online learning. Students also suggested improvements such as enhanced internet access, clearer instructional materials, and stronger teacher support. The study contributes to flexible learning literature by showing that students’ evaluations are shaped by learner autonomy, teacher presence, and access to technological resources, especially in a provincial teacher education context. Therefore, the findings highlight the need for improved instructional design and technological support to enhance flexible learning in teacher education.
The choice of a paramedical specialization suited to students’ academic profiles and preferences is a challenge for Moroccan students. To respond to this challenge, this study evaluated seven tree-based algorithms on real data from 889 nursing students from two Moroccan training institutes to determine the best-performing model to predict the students’ paramedical specialization. We implemented Bayesian hyperparameter optimization with Optuna, stratified 5-fold cross-validation, and final evaluation on an independent test set to ensure a robust and reliable evaluation of the models. The resampling techniques (SMOTE, SMOTEENN, ADASYN, SMOTETomek) were applied only in the training folds to avoid data leakage. Ensemble models (Random Forest, Extra Trees, XGBoost, Gradient Boosting, HistGradient Boosting) surpassed the baseline model (Decision Tree), while AdaBoost’s performance remained weak and unstable in cross-validation and on the test set (test macro F1-score as low as 0.1384), confirming its strong sensitivity to resampling-induced noise. On the test set, Random Forest with SMOTEENN achieved the best overall performance (macro F1-score = 0.8836, macro Receiver Operating Characteristic curve and Area Under the Curve (ROC-AUC) = 0.9879, macro Precision-Recall Area Under the Curve (PR-AUC) = 0.9294), whereas XGBoost with SMOTE delivered near-identical predictive performance (macro F1-score = 0.8828) with a faster inference time (0.0060 s), offering the best deployment trade-off. By integrating these two optimal approaches, an orientation system can accurately predict the specialization best suited to a student’s profile and preferences.
