<![CDATA[Reading ability is an important indicator of cognitive development in early childhood, particularly in kindergarten. This study aims to predict children’s reading readiness levels using the Backpropagation Neural Network (BPNN) algorithm. The data were obtained through observations and tests conducted on kindergarten children, with variables including age, letter recognition ability, phonemic ability, and concentration level. The BPNN model was trained by dividing the data into training and testing sets, using a single hidden layer and a sigmoid activation function. Model evaluation shows good predictive performance, with a Root Mean Squared Error (RMSE) of 0.527, indicating an average prediction error of less than 1% relative to the target values. These results confirm the ability of BPNN to recognize nonlinear patterns and accurately predict children’s reading readiness. Therefore, the application of BPNN can assist teachers and parents in designing appropriate learning interventions tailored to children’s developmental needs.]]>
Copyrights © 2026
