Next-Generation Question Creation: A Machine Learning Perspective on Document-Driven Automation
Dr.Sudhir Mohod, Arati Sawaj
Designing questions and answers is a necessary task when working for some organisations in the education and training sectors and even in some other fields with user interactions. However, the standard ways of doing it can be heavy on budget and resource. This work provides a brief description of a request online application built around AI to generate question and answer from text documents using recent NLP and machine learning techniques. Users that find the wizard interface of the web application (without having to register) dialogs very simple, will simply upload their text documents and get Questions and Answers in various formats including PDF, CSV and XLS. This paper presents a study on generating multiple-choice questions (MCQs) automatically quite distinctive from the fact that computer science education is a very evolving subject with multiple sub-domains.