The rapid growth of digital engagement among children has amplified both opportunities for learning and exposure to significant risks. While digital literacy (DL) equips children with essential competencies to navigate online environments, it does not inherently guarantee safety, particularly when mediation from parents or educators remains reactive rather than preventive. This study proposes a novel predictive framework integrating Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference Systems (ANFIS), and ensemble learning to model the interactions among DL, mediation, and risk. Exploratory statistical analysis revealed weak to moderate correlations, suggesting complex, nonlinear interdependencies that cannot be captured by linear models alone. PCA confirmed that risk and mediation strongly co-vary, whereas DL emerges as a distinct dimension. Baseline results showed that ANFIS outperformed ANN in handling fuzzy and uncertain relationships, achieving stronger correlations and lower error margins. However, ensemble approaches significantly improved predictive performance, with Ensemble-ANN achieving near-perfect accuracy (R = 0.999, RMSE < 0.0001), demonstrating its robustness and generalizability. The physical interpretation highlights DL as a double-edged sword, simultaneously empowering children while increasing exposure to risks, and underscores the need for mediation strategies that are proactive rather than reactive. Importantly, the findings align with international frameworks, including UNICEF’s Child Online Protection (COP), UNESCO’s Global Digital Literacy Framework, OECD’s socio-technical safeguards, and WHO’s digital health guidelines. This alignment reinforces the study’s contribution beyond computational advances, offering evidence-based insights for educators, parents, and policymakers seeking to balance empowerment and protection in children’s digital ecosystems.

Streamflow modelling is regarded as a crucial part of managing and planning water resources. Water resources engineers face a variety of challenges when predicting streamflow. These difficulties are caused by complex natural processes that involve non-linearity, non-stationarity, and randomness. This research investigates the application of machine learning (ML)-based models for forecasting streamflow discharge (Q) using input variables, including temperature and rainfall. The study attains the essential stationarity required for precise modelling using Augmented Dickey-Fuller tests, data normalization, and transformation. In contrast, unit root tests identify initial-level non-stationarity and call for first-differencing. Correlation matrix analysis identifies relevant input combinations. The result findings are supported by statistical metrics including mean squared error (MSE), mean absolute error (MAE), and Pearson correlation coefficient (PCC). Notably, in terms of prediction accuracy, the Gaussian Process Regression (GPR) GPR-M3 model stands out as a notable performer, with a low MAE value of 0.034 in the calibration phase and 0.027 in the verification phase. The success of these techniques is further supported by first and second-order ensemble algorithms, with some models reaching a perfect PCC score during both the calibration and verification phases. The study emphasizes the significance of preprocessing, model selection, and ensemble procedures in improving the accuracy of streamflow prediction models. In addressing complex nonlinear interactions, artificial intelligence (AI)- based models are valuable tools for both technical applications and practical understanding.