Author

Abeeha Sattar

Degree

Master of Science in Computer Science

Faculty / School

School of Mathematics and Computer Science (SMCS)

Department

Department of Computer Science

Date of Submission

2020-06-30

Advisor

Dr. Quratulain Rajput, Assistant Professor, Faculty of Computer Science, Institute of Business Administration (IBA), Karachi

Project Type

MSCS Survey Report

Abstract

Automatic Question Generation has become a popular topic among researchers in an educational context. Some academic institutes are using automatic question generation for assessing their students’ understanding of subjects. In this paper, we discuss about DBpedia, which contains the data from Wikipedia in a structured format. We will also be creating a prototype that generates questions based on entities extracted from DBpedia. We are using the Jena API and SPARQL to access DBpedia. The questions are created from templates based on different attributes of the entities. We conclude that using DBpedia as a means to gather attributes for question generation proves to be an efficient method to generate questions.

Notes

The purpose of this paper is to suggest a way to automatically generate questions based on Wikipedia articles. DBpedia contains the information from Wikipedia, in an organized and structured format. The goal of the prototype is to use this structured information to create questions based on question templates that are generated by people. For the sake of the research, a subset of the attributes was used and the questions were self-generated and not crowd-sourced. We have created a prototype that uses the attributes extracted from DBpedia to generate questions for a particular type of entity, i.e., Person. In this paper, we have discussed, outlined and demonstrated how to utilize DBpedia to generate questions for an entity. However, the demonstration is limited to only one type of entity. Future works can focus on extending the question generation process to other types of entities. Furthermore, the question templates can be crowd-sourced for better variations of questions.


The full text of this document is only accessible to authorized users.

Share

COinS