Automatic generation of multiple choice questions for e-assessment

It is important for students to expertise in their field of study, because there is an agile change in all the domains. Even though resources are available to learn, proper assessment helps them to improve upon their knowledge. In this paper, an automatic generation of multiple choice questions on any user-defined domain is proposed. It first extracts text relevant to the given domain from the web and summarises using fireflies-based preference learning. The sentences in the summary are transformed into stem for the MCQs. The distractors are generated using similarity metrics such as hypernyms and hyponyms. The system also generates analogy questions to test the verbal ability of the students.