Paper Title
Quiz Generator from Content, Books, or Articles Using Machine Learning

Abstract
The rapid increase in the use of digital education platform has generated a tremendous demand on automated and scalable means of assessment development. Traditional quiz creation uses a large amount of manual labor, topic knowledge, and pedagogical design, thus becoming cumbersome whenever generation applied to extensive or high production of learning resources. In order to address these drawbacks, this paper presents automated quiz generation system which can take assessment questions directly based on instructional text using Natural Language Processing methods as well as transformer-based models. The system uses text cleaning, key concepts extraction, and sentence selections followed by question generation using a fine-tuned text-to-text transfer transformer (T5). It has a modular design that allows converting unstructured learning material into structured and significant items in a quiz. According to the experimental findings, it has been shown that the questions produced are of high grammatical quality, semantically coherent and consistent with the original text. The proposed system will benefit the adaptive learning in contemporary learning settings since it will reduce human factor in the assessment design, thereby improving scalability of the system. This method is an effective and efficient way of automated generation of assessment in online learning. Keywords - Automatic Question Generation, Natural Language Processing, Transformer Models, Machine Learning, Educational Technology