DESIGN AND IMPLEMENTATION OF WEB BASED AI EDUCATIONAL ASSISTANT
Abstract
ABSTRACT
This project presents the design and implementation of a web based AIpowered educational assistant platform tailored for Computer Engineering
students. The system integrates four core Natural Language Processing (NLP)
capabilities; multiple-choice question generation using a fine-tuned BARTLarge model, question answering also usinga finetuned Flan-T5 model, text
summarization with BART-Large-CNN, and English–Hausa translation powered
by NLLB-200. These models are deployed as APIs and integrated into a userfriendly web application that enables students to upload learning materials,
receive automated quizzes, get answers to questions, and access concise
summaries or translations. The goal is to support personalized learning, selfassessment, and multilingual accessibility. The methodology includes model
fine-tuning, evaluation using metrics such as BLEU, ROUGE, and ChrF++, and
full-stack integration using FastAPI and a modern front-end framework. The
platform shows strong potential in enhancing educational accessibility and
intelligent content delivery, particularly for resource-constrained or multilingual
learning environments.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.