Application of Neural Networks in Question Generation

  • Himanshu Ashar, Devansh Mehta, Riddhi Shah, Neha Katre


In today’s time, when the world faces the issue of the raging pandemic of COVID-19, it is clear that substitute learning systems need to be put in place in order to ease the work of teachers, while continuing the education of students. In order to do this, self-sufficient tutoring systems need to be introduced to the world through the medium of a website or a mobile application. An important aspect of creating tutoring systems is the generation of questions based on the content of the subject which has to be taught. Traditionally, the process of test preparation is done manually by teachers and takes too much time and effort for creating multiple sets of equivalent questions to test the student’s knowledge about a particular topic. A notable problem to arise is gathering the questions from the external sources like authorized books, question banks which might not be relevant to the content which the student has learned. To facilitate the automation of this process, hence reducing the workload on teaching faculties in educational institutions, Neural Networks can be built and trained on the data gathered from above stated sources, to generate questions on various subjects of learning. This paper compares various techniques used to generate questions in an Intelligent Tutoring System.

How to Cite
Himanshu Ashar, Devansh Mehta, Riddhi Shah, Neha Katre. (2020). Application of Neural Networks in Question Generation. International Journal of Advanced Science and Technology, 29(3), 14265 - 14274. Retrieved from