An interactive system for teaching school going children.

Md Sahin Alam

Diplomová práce

Abstract:

Web-based interactive systems are becoming more common in educational systems for school-going children’s. Using AI, machine learning, and deep learning have huge potential in building such systems to improve personalised and adaptive learning experiences for these children. In this thesis paper I discuss and compare machine learning and deep learning algorithms to demonstrate that the Convoluted Neural Network is the most effective at recognising hand-written Bengali and English alphabets, digits and common shapes for improving accuracy of such interactive learning systems. From the results of this comparison I developed an interactive educational platform to improve student engagement, learning outcomes, and enjoyment of the learning process by adding a drawing tool as well as visual and auditory responses. The goal of this platform is to demonstrate the possibility of replacing traditional learning systems with digital interactive ones.

Keywords

Convolution Neural Networks, Discrimination of alphabet characters, Machine Learning, Scikit-Learn, Flask, TensorFlow, Panda, OpenCV

Jazyk práce: angličtina

Datum vytvoření / odevzdání či podání práce: 29. 3. 2023

Obhajoba závěrečné práce

• Obhajoba proběhla 22. 5. 2023
• Vedoucí: doc. Ing. Arnošt Veselý, CSc.
• Oponent: Richard Vágner, externi

Česká zemědělská univerzita v Praze

Provozně ekonomická fakulta

Magisterský studijní program:
Informatics