Key points are not available for this paper at this time.
An essential tool used by educational institutions to evaluate student performance is the examination process. The type of exam questions would dictate the caliber of students that the colleges produced. For the instructors, creating the exam questions is difficult, time-consuming, and unpleasant when they are new to the subject. For this reason, having an intelligent system that assesses students' learning abilities is crucial. We have proposed an automated question paper generation process that is quick, efficient, secure, and randomized. The exam paper is created and the inspection is completed by the design procedure using an efficient algorithm that has a high success rate. It is powered by deep learning and natural language processing (NLP) methods, particularly by using transformer models. With ease, the system can process a wide range of input formats, such as PDFs, audio files, and video, and transform them into a format that is standardized for efficient analysis. We have also tried to explore how to integrate Bloom's Taxonomy learning methods into the questions. Field-specific algorithms provide even more flexibility by guaranteeing that the question-creation process is in line with the particularities of each field which is important for subjects like mathematics. To provide a thorough assessment experience, the generator allows the construction of a wide range of question kinds, such as Fill-in-the-blanks (FIBs), Multiple-Choice Questions (MCQs), and Abstract question generation along with Mathematical word problems. Teachers benefit additionally from an automatically created model answer paper from the system, which makes paper correction quick and reliable.
Rathi et al. (Fri,) studied this question.