Leveraging a Hybrid Fuzzy C-Means-PCA Model for Identifying Speech Therapy Needs in Children
Journal of Data Science and Intelligent Systems · 05.06.2026 · 4 (4) · 445-458
DOI: 10.47852/bonviewjdsis62028311 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Data Science and Intelligent Systems, online erschienen am 05.06.2026. Verfasst von Muhammad Rizal Haris und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Kinderkrankenpflege
Abstract (Originalsprache)
This study proposes a hybrid unsupervised learning framework that integrates principal component analysis (PCA) and Fuzzy C-Means (FCM) to support early identification of speech therapy needs in children. The model utilizes secondary structured datasets containing acoustic–prosodic features such as MFCC, jitter, shimmer, harmonic-to-noise ratio, pitch, duration, fluency, and temporal indicators. PCA was applied to reduce dimensional redundancy and address the high-dimensional, low-sample-size characteristics of pediatric speech data, producing four principal components that retained 83.27% of the total variance. These components were subsequently clustered using FCM to capture partial membership patterns that reflect the continuous nature of children’s speech deviations. The proposed PCA-FCM model achieved the best cluster compactness and separation, with a 25.6% improvement in the XieBeni (XB) index compared to the baseline FCM model (XB = 0.421). Three interpretable clusters including Normal, Mild Deviation, and Severe Deviation were identified, each associated with distinct acoustic–prosodic profiles. These findings demonstrate the potential of hybrid unsupervised learning to provide an objective, interpretable, and efficient early-screening mechanism for guiding personalized speech therapy interventions in children. Received: 19 November 2025 | Revised: 16 April 2026 | Accepted: 11 May 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The analytical dataset supporting this study is publicly available on Figshare at https://doi.org/10.6084/M9.FIGSHARE.31410669. Author Contribution Statement Muhammad Rizal Haris: Conceptualization, Methodology, Validation, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Muhammad Faisal: Conceptualization, Methodology, Software, Formal analysis, Investigation, Resources, Data curation, Writing – review & editing, Supervision. Rizki Yusliana Bakti: Methodology, Software, Validation, Data curation, Project administration. Muhyiddin A. M Hayat: Validation, Formal analysis, Investigation, Resources, Project administration. Titik Khawa Abd Rahman: Conceptualization, Methodology, Software, Validation, Writing – review & editing, Project administration. Muhammad Syafaat S. Kuba: Validation, Writing – review & editing, Supervision. Titin Wahyuni: Data curation, Software, Project administration.
Zitieren
APA
Haris, M. R., Faisal, M., Bakti, R. Y., Hayat, M. A. M., Rahman, T. K. A., Kuba, M. S. S., & Wahyuni, T. (2026). Leveraging a Hybrid Fuzzy C-Means-PCA Model for Identifying Speech Therapy Needs in Children. Journal of Data Science and Intelligent Systems, 4(4), 445-458. https://doi.org/10.47852/bonviewjdsis62028311
Vancouver
Haris MR, Faisal M, Bakti RY, Hayat MAM, Rahman TKA, Kuba MSS, et al. Leveraging a Hybrid Fuzzy C-Means-PCA Model for Identifying Speech Therapy Needs in Children. Journal of Data Science and Intelligent Systems. 2026;4(4):445-458. doi:10.47852/bonviewjdsis62028311
BibTeX
@article{Haris2026,
author = {Haris, Muhammad Rizal and Faisal, Muhammad and Bakti, Rizki Yusliana and Hayat, Muhyiddin A. M and Rahman, Titik Khawa Abd and Kuba, Muhammad Syafaat S. and Wahyuni, Titin},
title = {Leveraging a Hybrid Fuzzy C-Means-PCA Model for Identifying Speech Therapy Needs in Children},
journal = {Journal of Data Science and Intelligent Systems},
year = {2026},
volume = {4},
number = {4},
pages = {445--458},
doi = {10.47852/bonviewjdsis62028311}
}
EUVIMED