<![CDATA[Stunting is a condition with long-term impacts on cognitive function and productivity, thus requiring accurate early detection to support effective interventions. This study aims to develop a machine learning–based stunting prediction application using the Support Vector Machine (SVM) algorithm, integrated into mobile and web-based systems. The development method follows the CRISP-DM framework to ensure a structured data mining process and incorporates the Behaviour Change Wheel (BCW) to encourage positive user behaviour change. A dataset of approximately 121,000 records obtained from Kaggle was used for model training. Evaluation using black-box testing demonstrated that the application achieved excellent system, information, and service quality, with a benefit score of 98%. The application was shown to improve diagnostic efficiency, enhance the accuracy of stunting risk identification, and support the acceleration of early intervention efforts in primary healthcare centres (Puskesmas). Its implementation is expected to contribute to reducing stunting prevalence and advancing the digital transformation of primary healthcare services. It is recommended that future development focus on optimising healthcare worker training, improving the user interface, and expanding system integration and service coverage to maximise application effectiveness.]]>
Copyrights © 2025
