Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection

In Indonesia, the stunting rate has reached 36%, significantly higher than the World Health Organization's (WHO) standard of 20%. This high prevalence underscores the urgent need for effective early detection methods. Traditional data mining approaches for stunting detection have primarily focu...

Full description

Saved in:
Bibliographic Details
Main Authors: Yohanes Setiawan, Mohammad Hamim Zajuli Al Faroby, Mochamad Nizar Palefi Ma’ady, I Made Wisnu Adi Sanjaya, Cisa Valentino Cahya Ramadhani
Format: Article
Language:English
Published: Department of Informatics, UIN Sunan Gunung Djati Bandung 2025-04-01
Series:JOIN: Jurnal Online Informatika
Subjects:
Online Access:https://join.if.uinsgd.ac.id/index.php/join/article/view/1495
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850223278394703872
author Yohanes Setiawan
Mohammad Hamim Zajuli Al Faroby
Mochamad Nizar Palefi Ma’ady
I Made Wisnu Adi Sanjaya
Cisa Valentino Cahya Ramadhani
author_facet Yohanes Setiawan
Mohammad Hamim Zajuli Al Faroby
Mochamad Nizar Palefi Ma’ady
I Made Wisnu Adi Sanjaya
Cisa Valentino Cahya Ramadhani
author_sort Yohanes Setiawan
collection DOAJ
description In Indonesia, the stunting rate has reached 36%, significantly higher than the World Health Organization's (WHO) standard of 20%. This high prevalence underscores the urgent need for effective early detection methods. Traditional data mining approaches for stunting detection have primarily focused on unimodal data, either tabular or image data alone, limiting the comprehensiveness and accuracy of the detection models. Modality-based modeling, which integrates image and tabular data, can provide a more holistic view and improve detection accuracy. This research aims to analyze modality-based modeling for the early detection of stunting. Two modalities, unimodal and multimodal, are used in this study. The main contributions of this research are the development of a comprehensive framework for modality-based analysis, the application of advanced data preprocessing techniques, and the comparison of various machine learning algorithms to identify the best model for stunting detection. The dataset, comprising images and tabular data, is sourced from Posyandu in Sidoarjo, Indonesia. Image data undergoes preprocessing, including background segmentation and feature extraction using the Gray Level Co-occurrence Matrix (GLCM), while tabular data is processed through categorical encoding. The Synthetic Minority Oversampling Technique (SMOTE) addresses class imbalance, and Principal Component Analysis (PCA) is used for dimensionality reduction. Unimodal modeling uses tabular or image data alone, while multimodal modeling combines both before classification. The study achieves the best F1 scores of 0.96, 0.91, and 0.90 for tabular-only, image-only, and image-tabular modalities, respectively, demonstrating the effectiveness of data balancing and dimensionality reduction techniques.
format Article
id doaj-art-bb9df0bdb29e4d44b49a93df51fe6f98
institution OA Journals
issn 2528-1682
2527-9165
language English
publishDate 2025-04-01
publisher Department of Informatics, UIN Sunan Gunung Djati Bandung
record_format Article
series JOIN: Jurnal Online Informatika
spelling doaj-art-bb9df0bdb29e4d44b49a93df51fe6f982025-08-20T02:06:01ZengDepartment of Informatics, UIN Sunan Gunung Djati BandungJOIN: Jurnal Online Informatika2528-16822527-91652025-04-01101536510.15575/join.v10i1.14951500Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting DetectionYohanes Setiawan0Mohammad Hamim Zajuli Al Faroby1https://orcid.org/0000-0001-6500-270XMochamad Nizar Palefi Ma’ady2I Made Wisnu Adi Sanjaya3Cisa Valentino Cahya Ramadhani4Department of Information Technology, Telkom University, Surabaya Campus, SurabayaDepartment of Data Science, Telkom University, Surabaya Campus, SurabayaDepartment of Information Systems, Telkom University, Surabaya Campus, SurabayaDepartment of Data Science, Telkom University, Surabaya Campus, SurabayaDepartment of Information Technology, Telkom University, Surabaya Campus, SurabayaIn Indonesia, the stunting rate has reached 36%, significantly higher than the World Health Organization's (WHO) standard of 20%. This high prevalence underscores the urgent need for effective early detection methods. Traditional data mining approaches for stunting detection have primarily focused on unimodal data, either tabular or image data alone, limiting the comprehensiveness and accuracy of the detection models. Modality-based modeling, which integrates image and tabular data, can provide a more holistic view and improve detection accuracy. This research aims to analyze modality-based modeling for the early detection of stunting. Two modalities, unimodal and multimodal, are used in this study. The main contributions of this research are the development of a comprehensive framework for modality-based analysis, the application of advanced data preprocessing techniques, and the comparison of various machine learning algorithms to identify the best model for stunting detection. The dataset, comprising images and tabular data, is sourced from Posyandu in Sidoarjo, Indonesia. Image data undergoes preprocessing, including background segmentation and feature extraction using the Gray Level Co-occurrence Matrix (GLCM), while tabular data is processed through categorical encoding. The Synthetic Minority Oversampling Technique (SMOTE) addresses class imbalance, and Principal Component Analysis (PCA) is used for dimensionality reduction. Unimodal modeling uses tabular or image data alone, while multimodal modeling combines both before classification. The study achieves the best F1 scores of 0.96, 0.91, and 0.90 for tabular-only, image-only, and image-tabular modalities, respectively, demonstrating the effectiveness of data balancing and dimensionality reduction techniques.https://join.if.uinsgd.ac.id/index.php/join/article/view/1495data balancingdimensionality reductionmultimodalstuntingunimodal
spellingShingle Yohanes Setiawan
Mohammad Hamim Zajuli Al Faroby
Mochamad Nizar Palefi Ma’ady
I Made Wisnu Adi Sanjaya
Cisa Valentino Cahya Ramadhani
Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection
JOIN: Jurnal Online Informatika
data balancing
dimensionality reduction
multimodal
stunting
unimodal
title Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection
title_full Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection
title_fullStr Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection
title_full_unstemmed Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection
title_short Modality-based Modeling with Data Balancing and Dimensionality Reduction for Early Stunting Detection
title_sort modality based modeling with data balancing and dimensionality reduction for early stunting detection
topic data balancing
dimensionality reduction
multimodal
stunting
unimodal
url https://join.if.uinsgd.ac.id/index.php/join/article/view/1495
work_keys_str_mv AT yohanessetiawan modalitybasedmodelingwithdatabalancinganddimensionalityreductionforearlystuntingdetection
AT mohammadhamimzajulialfaroby modalitybasedmodelingwithdatabalancinganddimensionalityreductionforearlystuntingdetection
AT mochamadnizarpalefimaady modalitybasedmodelingwithdatabalancinganddimensionalityreductionforearlystuntingdetection
AT imadewisnuadisanjaya modalitybasedmodelingwithdatabalancinganddimensionalityreductionforearlystuntingdetection
AT cisavalentinocahyaramadhani modalitybasedmodelingwithdatabalancinganddimensionalityreductionforearlystuntingdetection