Domingo, Sophia Kaye G.; Dominguez, Clarissa V.; Espeño, Stephen E.
ABCiFy: Enhancing Alphabet Learning through image classification using convolutional neural network - Capstone Project: (Bachelor of Science in Information Technology) - Pamantasan ng Lungsod ng Maynila, 2025
ABSTRACT: Traditional methods of teaching the alphabet often struggle to maintain young children’s engagement and typically lack auditory and interactive assessment features, thereby limiting their effectiveness in supporting simultaneous learning. This study introduces ABCify, a web-based alphabet leaning system designed for children aged three to six. ABCiFy integrates Convolutional Neural Network (CNN) technology for image classification, Text-to-Speech (TTS) functionality to aid pronunciation, and an interactive quiz component for real-time assessment and reinforcement. Developed using the Agile methodology, the system was iteratively refined to align with defined functional requirements. Evaluation results indicate that the CNN model achieved an accuracy of 85% in object classification and alphabet association. The TTS feature effectively supported pronunciation learning, while the quiz component facilitated the validation of children’s progress in alphabet recognition. These findings suggest that ABCiFy has the potential to transform traditional alphabet learning into a more engaging, multi-sensory educational experience. Future work will focus on enhancing the CNN model’s accuracy, expanding the object database to support broader learning contexts, implementing a color-coded indicator in the Progress Tracker to visually represent performance levels, and incorporating visual records of quiz responses to promote reflective learning and error analysis.
academic writing
T58.6 D66 2025
ABCiFy: Enhancing Alphabet Learning through image classification using convolutional neural network - Capstone Project: (Bachelor of Science in Information Technology) - Pamantasan ng Lungsod ng Maynila, 2025
ABSTRACT: Traditional methods of teaching the alphabet often struggle to maintain young children’s engagement and typically lack auditory and interactive assessment features, thereby limiting their effectiveness in supporting simultaneous learning. This study introduces ABCify, a web-based alphabet leaning system designed for children aged three to six. ABCiFy integrates Convolutional Neural Network (CNN) technology for image classification, Text-to-Speech (TTS) functionality to aid pronunciation, and an interactive quiz component for real-time assessment and reinforcement. Developed using the Agile methodology, the system was iteratively refined to align with defined functional requirements. Evaluation results indicate that the CNN model achieved an accuracy of 85% in object classification and alphabet association. The TTS feature effectively supported pronunciation learning, while the quiz component facilitated the validation of children’s progress in alphabet recognition. These findings suggest that ABCiFy has the potential to transform traditional alphabet learning into a more engaging, multi-sensory educational experience. Future work will focus on enhancing the CNN model’s accuracy, expanding the object database to support broader learning contexts, implementing a color-coded indicator in the Progress Tracker to visually represent performance levels, and incorporating visual records of quiz responses to promote reflective learning and error analysis.
academic writing
T58.6 D66 2025