Optimizing Deep Learning with Integrated Data Balancing Strategies for Multiclass Stunting Prediction

Wulan Sri Lestari; Caroline; Mustika Ulina

doi:10.56741/IISTR.esl.002311

Authors

Wulan Sri Lestari
wulan.lestari@mikroskil.ac.id
Universitas Mikroskil https://orcid.org/0009-0004-5991-6926
Caroline Universitas Mikroskil
Mustika Ulina Universitas Mikroskil https://orcid.org/0009-0003-6065-3994

Keywords:

Deep Neural Networks, Imbalanced Data Handling, Multiclass Classification, SMOTE-NC, Stunting Classification

Abstract

Accurate prediction of stunting among toddlers plays a vital role in enabling early intervention and supporting data-driven public health planning. Nonetheless, constructing robust predictive models remains a challenge due to significant class imbalance and heterogeneous data characteristics derived from multiple sources. This research proposes an enhanced Deep Neural Network (DNN) model for multiclass stunting classification by integrating advanced data balancing (SMOTE, SMOTE-NC, and SMOTE-ENN). The model was trained using anthropometric data obtained from both primary and secondary datasets. Experimental evaluations indicate that the combination of the DNN and SMOTE-ENN model yielded optimal performance, obtaining an accuracy of 98.57%, precision of 98.59%, F1-Score of 98.58%, and ROC-AUC of 99.97%. The confusion matrix analysis showed a similar misclassification pattern, predominantly between the Severely Stunted and Stunted classes and between the Normal class and the adjacent growth-status classes, which suggests a considerable overlap of features between adjacent classes. Furthermore, the smooth training and validation loss curves show that the proposed model can effectively avoid the underfitting and overfitting problems, guaranteeing the reliability and robustness of the prediction. Overall, the combination of DNN+SMOTE-ENN was the best approach to deal with class imbalance, enhance class separation, and boost the performance of multiclass stunting classification.

Downloads

Download data is not yet available.

Author Biographies

Wulan Sri Lestari, Universitas Mikroskil

Information Technology

Caroline, Universitas Mikroskil

Information System

Mustika Ulina, Universitas Mikroskil

Information Technology

References

J. R. Khan, J. H. Tomal, and E. Raheem, “Model and Variable Selection using Machine Learning Methods with Applications to Childhood Stunting in Bangladesh,” Informatics for Health and Social Care, Vol. 46, No. 4, pp. 425-442, 2021, doi: https://doi.org/10.1080/17538157.2021.1904938.

M. de Onis et al., “Prevalence thresholds for wasting, overweight and stunting in children under five years,” Public Health Nutrition, vol. 22, no. 1, pp. 175–179, Jan. 2019, doi: https://doi.org/10.1017/s1368980018002434

C. D. Bourke, K. D. Jones, and A. J. Prendergast, “Current Understanding of Innate Immune Cell Dysfunction in Childhood Undernutrition,” Frontiers in Immunology, vol. 10, p. 1728, Jul. 2019, doi: https://doi.org/10.3389/fimmu.2019.01728.

H. Pohan, M. Zarlis, E. Irawan, H. Okprana, and Y. Pranayama, “Clustering stunted toddlers in Indonesia using the K-Medoids algorithm,” Public Knowledge Project PLN, Vol. 3, No. 2, pp. 97–104, 2021, doi: https://doi.org/10.53842/juki.v3i2.69.

J. V. Moniaga, M. Ohyver, J. Siregar, and P. H. Yauwito, “Map-type Modelling and Analysis of Children Stunting Case Data in Indonesia with Interactive Multimedia Method,” Procedia Computer Science, vol. 157, pp. 530–536, 2019, doi: https://doi.org/10.1016/j.procs.2019.09.010.

S. Lonang and D. Normawati, “Classification of stunting status in toddlers using K-Nearest Neighbor with backward elimination feature selection,” Journal Media Informatika Budidarma, Vol. 6, No. 1, pp. 49–49, 2022, doi: https://doi.org/10.30865/mib.v6i1.3312.

M. Rosyidah, Y. L. R. Dewi, and I. Qadrijati, “Effects of Stunting on Child Development: A Meta-Analysis,” Journal of Maternal Child Health, Vol. 6, No. 1, 2021, https://thejmch.com/index.php/thejmch/article/view/522.

S. Zaleha and H. Idris, “Implementation of Stunting Program in Indonesia: A Narrative Review,” Indonesian Journal of Health Administration, Vol. 10, No. 1, pp. 143–151, 2022, doi: https://doi.org/10.20473/jaki.v10i1.2022.143-151.

B. R. Bait et al., “Scaling up child wasting prevention and treatment in the context of stunting prevention in Indonesia,” https://www.ennonline.net/fex/67/en/scaling-child-wasting-prevention-and-treatment-context-stunting-prevention-indonesia (accessed: 2025)

R. Resmiati and T. Arifin, “Classification of Breast Cancer Patients Using the Support Vector Machine Method with Backward Elimination,” SISTEMASI, Vol. 10, No. 2, pp. 381-393, 2021, doi: https://doi.org/10.32520/stmsi.v10i2.1238.

I. Ayuningtyas and E. U. Kasanah, “Application of SMOTE for analyzing COVID-19’s impact on working-age population in East Kalimantan,” BESTARI BPS Kalimantan Timur, Vol. 1, No. 1, 2021, https://bestari.bpskaltim.com/index.php/bestari-bpskaltim/article/view/3

O. N. Chilyabanyama et al., “Performance of Machine Learning Classifiers in Classifying Stunting among Under-Five Children in Zambia,” Children, Vol. 9, No. 7, 2022, doi: https://doi.org/10.3390/children9071082.

R. Syahrial, R. Ilham, Z. F. Asikin, and S. Surya, “Stunting Classification in Children's Measurement Data Using Machine Learning Models,” Journal La Multiapp, Vol. 3, No. 2, pp. 52–60, 2022, doi: https://doi.org/10.37899/journallamultiapp.v3i2.614.

S. Y. Andriyani, M. S. Lydia, and S. Efendi, “Optimization of Support Vector Machine Algorithm Using Stunting Data Classification,” Prisma Sains, Vol. 11, No. 1, pp. 164–171, 2023, doi: https://doi.org/10.33394/j-ps.v11i1.6619.

F. M. Amin and D. C. R. Novitasari, “Identification of Stunting Disease using Anthropometry Data and Long Short-Term Memory (LSTM) Model,” Comput. Eng. Appl. J., Vol. 11, No. 1, pp. 25–36, 2022, doi: https://doi.org/10.18495/comengapp.v11i1.395.

W. S. Lestari, C. Caroline, and Y. M. Saragih, “Deep Learning Approach for Stunting Classification in Toddlers,” Proc. 2nd Int. Conf. Technol. Innov. Appl. (ICTIIA), Medan, Indonesia, 2024, pp. 1–5, doi: https://doi.org/10.1109/ICTIIA61827.2024.10761797.

C. J. Ananta, A. Fariza, and R. Asmara, “Stunting Program Classification in East Java, Indonesia From Internet News Using Location-Based and SVM,” Proc. Int. Electron. Symp. (IES), pp. 527–532, 2023, doi: https://doi.org/10.1109/IES59143.2023.10242418.

E. Sutoyo and M. A. Fadlurrahman, “Application of SMOTE to Overcome Class Imbalance in Television Advertisement Performance Rating Classification Using Artificial Neural Network,” J. Edukasi Penelit. Informatika (JEPIN), Vol. 6, No. 3, pp. 379–385, 2020, doi: https://doi.org/10.26418/jp.v6i3.42896.

S. Yadav, “SMOTE in Predictive Modeling: A Comprehensive Evaluation of Synthetic Oversampling for Class Imbalance,” Int. J. Innov. Res. Eng. Multidiscip. Phys. Sci., Vol. 8, No. 4, 2020, doi: https://doi.org/10.5281/zenodo.14259555.

T. Sugihartono, B. Wijaya, M. Marini, A. P. Alkayess, and H. A. Anugerah, “Optimizing Stunting Detection through SMOTE and Machine Learning: a Comparative Study of XGBoost, Random Forest, SVM, and k-NN,” J. Appl. Data Sci., Vol. 6, No. 1, pp. 667–682, 2025, https://doi.org/10.47738/jads.v6i1.494.

E. Miranda, M. Aryuni, A. Y. Zakiyyah, Y. E. Kurniawati, A. V. D. Sano, and M. Kumbangsila, “An early prediction model for toddler nutrition based on machine learning from imbalanced data,” Procedia Computer Science, vol. 245, pp. 263–271, 2024, doi: https://doi.org/10.1016/j.procs.2024.10.251.

W. M. Baihaqi, I. N. Laela, and D Darso, “Improving Stunting Classification Performance Using Combination SMOTE Technique and Artificial Neural Network Algorithm,” JITK, Vol. 10, No. 1, pp. 160–167, 2024, doi: https://doi.org/10.33480/jitk.v10i1.4998.

J. Minardi, F. Budiman, M. Z. Fanani, and N. Rijati, “Enhancing Stunting Detection Accuracy in Children Using SVM with Advanced Data Balancing Techniques,” J. Inf. Syst. Eng. Manage., Vol. 10, No. 6s, pp. 501–509, 2025, doi: https://doi.org/10.52783/jisem.v10i6s.748.

G. Ras, N. Xie, M. V. Gerven, and D. Doran, “Explainable Deep Learning: A Field Guide for the Uninitiated,” Journal of Artificial Intelligence Research, Vol. 73, pp. 329–397, 2022, doi: https://doi.org/10.1613/jair.1.13200.

W. S. Lestari and M. Ulina, “Optimizing Deep Neural Networks Using ANOVA for Phishing Detection,” Teknika, Vol. 13, No. 1, pp. 71–76, 2024, 76, Feb. 2024, doi: https://doi.org/10.34148/teknika.v13i1.758.

S. Fouladi, A. A. Safaei, N. Mammone, F. Ghaderi, and M. J. Ebadi, “Efficient Deep Neural Networks for Classification of Alzheimer’s Disease and Mild Cognitive Impairment from Scalp EEG Recordings,” Cognitive Computation, Vol. 14, pp. 1247-1268, 2022, doi: https://doi.org/10.1007/s12559-022-10033-3.

N. K. Trivedi et al., “Early Detection and Classification of Tomato Leaf Disease Using High-Performance Deep Neural Network,” Sensors, Vol. 21, No. 23, p. 7987, 2021, doi: https://doi.org/10.3390/s21237987.

N. Mubarakah, P. Sihombing, S. Efendi, and F. Fahmi, “Target Image Validation Modeling Using Deep Neural Network Algorithm,” International Journal of Electrical and Computer Engineering (IJECE), Vol. 15, No. 2, pp. 2042–2054, 2025, doi: http://doi.org/10.11591/ijece.v15i2.pp2042-2054.

J. H. Joloudari, A. Marefat, M. A. Nematollahi, S. S. Oyelere, and S. Hussain, “Effective Class-Imbalance Learning Based on SMOTE and Convolutional Neural Networks,” Appl. Sci., Vol. 13, No. 6, 2023, doi: https://doi.org/10.3390/app13064006.

R. P. Pradana, “Stunting Toddler (Balita) Detection (121K rows),” Kaggle.com, 2024. https://www.kaggle.com/datasets/rendiputra/stunting-balita-detection-121k-rows (accessed 2025).