Quiz Maker: Automatic Quiz Generation from Text Using NLP

Abstract

In the past few years, there has been great technological advancement in the field of deep learning and natural language processing. One of the applications is automatic generation of quizzes from text. The recent advancement in NLP techniques has shown a lot of promise. The proposed solution uses an NLP pipeline involving Bert and T5 transformers to extract keywords and gain insights from the text input. From the extracted keywords, different types of questions are generated such as fill in the blanks, true or false, Wh-type and multiple choice questions. Latest state-of-the-art models proved to perform better in all stages of our pipeline. The results from these models have shown a lot of promise. Through a survey created for evaluating the model, around 60% questions generated by the model were incorrectly identified as human generated or could not be determined by the survey participants.