ARCHIVES

In the evolving domain of digital education, the need for personalized, interactive, and intelligent academic support has become increasingly significant. Traditional e-learning platforms often lack the ability to provide real-time, adaptive responses to student queries, thereby creating learning gaps and reducing engagement. This study proposes an AI-Enhanced Educational Assistant designed to deliver dynamic, multimodal support for learners through real-time question answering, emotional state recognition, and personalized content generation. The framework integrates advanced Natural Language Processing (NLP) models such as GPT and BERT with Computer Vision (CV) techniques for image-based query handling and Convolutional Neural Networks (CNNs) for emotion recognition. A multimodal interaction pipeline enables seamless communication via text, voice, and image, while reinforcement learning with user feedback ensures continuous system improvement. The assistant also supports additional features including quiz generation, assignment guidance, progress tracking, and integration with learning management systems (LMS). The system was implemented with Python, TensorFlow, FastAPI, and Streamlit, and evaluated through benchmark datasets and student feedback. Results demonstrate that the assistant not only improves academic engagement and retention but also enhances inclusivity by adapting to learners’ cognitive and emotional states. This work highlights the potential of AI-driven educational technologies to bridge gaps in remote and hybrid learning environments, offering a scalable and empathetic digital learning companion.