AI-Driven Ethnoscience Learning: Enhancing Physics Education Through Malay Cultural Insights
, M Furqon, Khairul Anwar, Hebat Shidow Falah, Cicyn RiantoniThis study aims to explore the innovative integration of machine learning techniques into project-based learning rooted in Malay ethnoscience in Jambi, Indonesia. The research introduces a novel framework that utilizes educational data mining to personalize culturally responsive STEM education in under-resourced public schools.
Ethnoscience, particularly ethnophysics, provides a culturally relevant context for learning, connecting scientific principles to traditional practices. In Jambi, rich traditions such as the construction of stilt houses and the preparation of local cuisine reflect underlying principles of physics and sustainability. Integrating these practices into education is crucial for enhancing students’ critical thinking and contextual understanding.
This research adopts a mixed-methods approach, combining qualitative analysis of Malay ethnoscience practices with data mining techniques to extract patterns and correlations from educational data. Project-based learning modules were designed based on insights derived from data mining algorithms applied to student performance and engagement metrics.
The study contributes a unique model combining cultural heritage with predictive analytics to optimize learning design. Unlike existing AI-driven systems, this model embeds local wisdom as a predictive variable, enabling contextualized learning recommendations grounded in students’ sociocultural identities.
Key findings include a 25% increase in student engagement and a 15–20% improvement in understanding physics concepts among students engaged in culturally embedded hands-on projects. Machine learning algorithms (Random Forest and Naïve Bayes) achieved up to 85% accuracy in predicting student success, identifying prior cultural familiarity and level of active engagement as the most influential predictors.
Educators are encouraged to use data mining tools to personalize learning and incorporate local wisdom into curricula. Culturally rooted project-based learning has proven effective in enhancing student engagement.
Further studies are needed to expand the application of data mining in ethnoscience education across other cultural contexts. Researchers should explore the scalability of such frameworks and the integration of additional machine learning algorithms to enhance prediction accuracy.
By bridging traditional knowledge with modern AI, this study offers a scalable model for culturally inclusive education. It demonstrates how local wisdom can be leveraged through machine learning to promote both academic excellence and cultural preservation, particularly in public education systems facing resource constraints.
Future work should focus on the development of interactive tools and digital resources to support project-based ethnoscience learning. Studies should also investigate the long-term impact of integrating ethnoscience into formal education on cultural preservation and academic performance.
Back