<![CDATA[Considering the maternal knowledge, economic status, and maternal nutritional status, the current study created an optimal risk assessment model to detect childhood stunting risk. At the same time, these variables are unbalanced and interrelated in non-linear fashion. Then, to these ends, an Ensemble Bagging model consisting of Support Vector Machine and Kernel Discriminant was trained by voting on the aggregation of the majority of 100 bootstrapped samples, which countered variance and overfitting reducing, hence improving generalization. The primary data were sourced from the mothers of toddlers in the Wajak District. The model predicators were 3 out of the primary ones accounting for the stunting risk. The model also recorded an accuracy of 95%, sensitivity level of 80%, as well as a 100% specificity score. Non-linear relationships were detected and the variance was also reduced, supporting the study to place itself in the realms of novelty by being the first research to fuse the Ensemble Bagging with Kernel methods for Detected stunting risk, The model, hence, fits best as a decision Support System for detecting stunting risk at an early stage.]]>
Copyrights © 2025
