Ensembled DNN Model Based Enhancing Quality Education: Evaluating the Impact of Mindfulness of Students

Attention and concentration are critical cognitive skills that underpin academic success and overall child development. This study evaluates the impact of mindfulness-based interventions (MBIs) on attention and concentration in school-age children (6–12 years) using an ensembled deep neural network (DNN) model. Participants were stratified by age, gender, and prior mindfulness exposure, ensuring diversity and representation across socio-economic and geographic backgrounds. The MBI program included guided meditation, breathing exercises, and mindful activities delivered over 8–12 weeks, facilitated by certified instructors. Pre- and post-intervention assessments were conducted using standardized tools like the Stroop Test and Continuous Performance Task, supplemented by self-reports, surveys, and behavioral observations. The ensembled DNN model, combining Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs), achieved an accuracy of 98.11% on the training set, 97.85% on the validation set, and 97.50% on the testing set. Significant improvements were observed in attention metrics, including response time reduction (from 720ms to 655ms) and increased Stroop Test accuracy (82% to 89%, p< 0.001). Behavioral observations and qualitative feedback corroborated these findings, highlighting increased focus, reduced distractibility, and improved emotional regulation. This study demonstrates the effectiveness of MBIs in enhancing cognitive and emotional skills in children and validates the use of AI models for scalable evaluation. Future research should explore longitudinal effects and adaptability to varied cultural and educational contexts.