JURNAL EDUKASI DAN PEMBELAJARAN ANAK USIA DINI
Vol. 2 No. 2 (2026)

Application Of Deep Learning Model In Improving Early Childhood Development

Meyzia Susang (Institut Agama Kristen Negeri Kupang)
Yelsi Benu (Institut Agama Kristen Negeri Kupang)
Rosalia Blegur (Institut Agama Kristen Negeri Kupang)
Marina Anone (Institut Agama Kristen Negeri Kupang)
Sarita Natumnea (Institut Agama Kristen Negeri Kupang)
Selvi Wahi (Institut Agama Kristen Negeri Kupang)
Fredericksen Victoranto Amseke (Institut Agama Kristen Negeri Kupang)



Article Info

Publish Date
15 Jun 2026

Abstract

Background: Cognitive development is an essential aspect of early childhood education as it underpins children’s abilities in thinking, reasoning, problem-solving, and decision-making that support lifelong learning. The Deep Learning model, which integrates mindful, meaningful, and joyful learning principles, is an innovative approach designed to enhance children’s engagement and learning experiences. However, empirical evidence on its effectiveness in relation to cognitive development in real kindergarten settings remains limited. Objective: This study aimed to analyze the relationship between the implementation of the Deep Learning model and the cognitive development of children aged 4–6 years at Cemara Liliba Kindergarten, Kupang City. Method: This quantitative study used an ex post facto design involving 20 children selected through purposive sampling. Data were collected through observation, interviews, documentation, and cognitive development assessments, then analyzed using descriptive statistics and simple linear regression with SPSS. Results: The findings indicated that the Deep Learning model encouraged active participation, curiosity, and enjoyable learning experiences among children through mindful, meaningful, and joyful activities. However, the regression analysis showed that the relationship between the Deep Learning model and cognitive development was not statistically significant (p > 0.05). The coefficient of determination (R² = 0.088) revealed that only 8.8% of cognitive development variance was explained by the model, while 91.2% was influenced by other factors. Novelty: This study provides empirical evidence of Deep Learning implementation in early childhood classrooms using an ex post facto design. Conclusion: The Deep Learning model created positive learning experiences but did not significantly affect children’s cognitive development.

Copyrights © 2026






Journal Info

Abbrev

E-MAS

Publisher

Subject

Education Other

Description

Jurnal Edukasi dan Pembelajaran Anak Usia Dini memuat publikasi ilmiah hasil penelitian, studi kepustakaan, pengembangan model, dan praktik baik dalam bidang pendidikan dan pembelajaran anak usia dini. Fokus utama jurnal ini mencakup dua ranah utama: (1) edukasi secara umum di berbagai jenjang, dan ...