Hybrid Machine Learning to Evaluate the Incidence of Toddler Stunting through Integration of Multi-source Satellite Imagery and Official Statistics in East Nusa Tenggara Province
Abstract
Stunting is a serious health problem that impacts the quality of life of children under five. In 2023, East Nusa Tenggara recorded the second highest prevalence of stunting in Indonesia, influenced by health, socio-economic and environmental factors. In terms of the environment, remote sensing technology can be utilised to monitor environmental factors that contribute to stunting, such as vegetation conditions, access to clean water, and soil conditions. This study aims to evaluate the incidence of stunting among children under five using a hybrid machine learning approach, combining predictive modeling and cluster analysis. The results indicate thStunting is a serious health problem that impacts the quality of life of children under five. In 2023, East Nusa Tenggara recorded the second highest prevalence of stunting in Indonesia, influenced by health, socio-economic and environmental factors. In terms of the environment, remote sensing technology can be utilised to monitor environmental factors that contribute to stunting, such as vegetation conditions, access to clean water, and soil conditions. This study aims to evaluate the incidence of stunting among children under five using a hybrid machine learning approach, combining predictive modeling and cluster analysis. The results indicate that eXtreme Gradient Boosting Regressor (XGBR) is the best model for estimating stunting prevalence, with a Root Mean Squared Error (RMSE) of 3.2076 and an value of 0.7223. Meanwhile, for clustering results, K-Means Clustering is identified as the most effective method for grouping districts/cities based on socioeconomic and environmental factors. The clustering process produced two groups, such as vulnerable (Cluster 1) and highly vulnerable (Cluster 2), with connectivity, Dunn Index, and silhouette coefficient values of 29.290, 0.6931, and 0.4509, respectively. These findings are expected to serve as a basis for policymakers in formulating targeted interventions to reduce stunting rates, particularly in highly vulnerable areas. at eXtreme Gradient Boosting Regressor (XGBR) is the best model for estimating stunting prevalence, with a Root Mean Squared Error (RMSE) of 3.2076 and an value of 0.7223. Meanwhile, for clustering results, K-Means Clustering is identified as the most effective method for grouping districts/cities based on socioeconomic and environmental factors. The clustering process produced two groups, such as vulnerable (Cluster 1) and highly vulnerable (Cluster 2), with connectivity, Dunn Index, and silhouette coefficient values of 29.290, 0.6931, and 0.4509, respectively. These findings are expected to serve as a basis for policymakers in formulating targeted interventions to reduce stunting rates, particularly in highly vulnerable areas.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
