Random Search Optimization of Hyperparameter in Random Forest Algorithm for Stunting Prediction

Stunting, a condition of impaired growth in children due to prolonged malnutrition, remains a critical health issue in Indonesia, with a prevalence of 24.4% in 2021. Early detection is essential to mitigate its long-term effects, including cognitive deficits and chronic disease risks. This study leverages the Random Forest algorithm, enhanced with Random Search Optimization, to predict stunting in children. By optimizing hyperparameters such as the number of trees, maximum tree depth, and minimum samples per leaf, the model achieves high performance on a 70:30 data split, with an accuracy of 0.998, precision of 1, recall of 0.998, and F1-score of 0.998. This approach demonstrates the potential for accurate and timely stunting prediction, supporting targeted interventions to combat this pressing health challenge.