Comparative analysis of Machinegenerated questions (Quillionz) and Human-generated questions

University essay from KTH/Skolan för elektroteknik och datavetenskap (EECS)

Author: Mazen Bahy; [2020]

Keywords: ;

Abstract: This thesis investigates the quality of machine-generated questions compared to human-generated questions in terms of the grammar and the clarity of the questions. Limited work has previously been done in the area of comparing them in terms of the grammar and the clarity using different statistic methods to analyze the results. This thesis used machine-generated questions from Quillionz and human-written questions. The type of generated questions was Wh-questions. A survey was conducted and shared over social media to test if the people can determine human- from machine -generated questions and tell us their opinions about the grammar and the clarity of the questions. The results showed that machine- and human- generated questions have almost the same level of quality when it comes to grammar and clarity. At the same time, it showed that people could determine the type of question about 46% of the time.

  AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)