While exam-style questions are a fundamental educational tool serving a variety of purposes, manual construction of questions is a complex process that requires training, experience, and resources. This, in turn, hinders and slows down the use of educational activities (e.g. providing practice questions) and new advances (e.g. adaptive testing) that require a large pool of questions. To reduce the expenses associated with manual construction of questions and to satisfy the need for a continuous supply of new questions, automatic question generation (AQG) techniques were introduced. This review extends a previous review on AQG literature that has been published up to late 2014. It includes 93 papers that were between 2015 and early 2019 and tackle the automatic generation of questions for educational purposes. The aims of this review are to: provide an overview of the AQG community and its activities, summarise the current trends and advances in AQG, highlight the changes that the area has undergone in the recent years, and suggest areas for improvement and future opportunities for AQG. Similar to what was found previously, there is little focus in the current literature on generating questions of controlled difficulty, enriching question forms and structures, automating template construction, improving presentation, and generating feedback. Our findings also suggest the need to further improve experimental reporting, harmonise evaluation metrics, and investigate other evaluation methods that are more feasible.

/pdf/a-systematic-review-of-automatic-question-generation-for-4or9ww94ie.pdf

A Systematic Review of Automatic Question Generation for Educational Purposes

In this paper, we present the system of automatic MCQs (Multiple Choice Questions) generation for any given input text along with a set of distractors. The system is trained on a Wikipedia-based dataset consisting of URLs of Wikipedia articles. The important words (keywords) which consist of both bigrams and unigrams are extracted and stored in a dictionary along with many other components of the knowledge base. We have used Inverse Document Frequency (IDF) measure for ranking the extracted keywords and Context-Based Similarity approach using Paradigmatic Relation discovery techniques for generation of distractors. In addition, the question generation phase includes eliminating sentences starting with Discourse Connectives to avoid a question with incomplete information. We have obtained significant accuracy compared to many similar approaches. The results are quite promising considering that there is no human intervention. The task of automatic question generation can be of quite an importance in MOOCs (Massive Open Online Courses), Intelligent Tutoring Systems and for self-assessment while learning a new concept. Though we have developed our system for the field of physics, it can be extended to any field.

Automatic question generation for intelligent tutoring systems

Abstract This paper presents an environment which generates tests automatically. It is designed for assistance in the software engineering education and is part of the Virtual Education Space. The environment has two functionalities – generation and assessment of different types of test questions. In the paper, the architecture of the environment is described in detail. The test generation is supported by specialized ontologies, which are served by two intelligent agents known as Questioner Operative and Assessment Operative.

/pdf/an-environment-for-automatic-test-generation-2caziixvai.pdf

An Environment for Automatic Test Generation

The invention relates to a test question structuring method which comprises the following two stages: (1), the preparation stage of test questions in a text form: inputting the test questions in the text form according to the prescribed format, and not adding additional auxiliary information; (2), the automatic computer identification stage: carrying out automatic identification on all the test questions in the text form, which are input according to the prescribed format in the stage (1), according to judgment rules and processing modes, and outputting structured test questions. According to the method, structured data processing is completed through automatic computer identification, all needed is to input the test questions in the text form according to the prescribed format in the preparation stage of the test questions in the text form manually, and the additional auxiliary information is not required to be added for the test questions. The prescribed format is basically consistent with the format of the test questions in a paper form generally used at present, so that too much additional work is avoided in the preparation stage and both time and labor are saved.

Test question structuring method

Air Force Office of Scientific Research, Office of Naval Research, National Science Foundation

The Genesis Enterprise: Taking Artificial Intelligence to another Level via a Computational Account of Human Story Understanding

For competitive examinations (e.g. GMAT, CAT) and for various offline and online academic courses, test questions need to be written by human authors. These question-writing tasks take significant manual effort and time. To simplify this effort, we propose an Automatic Question Generation system that focuses on questions that can be generated in English from structured data. This paper explains an approach, which takes a data table as input and generates questions of varying types. The data is preprocessed and categorized according to their class (numeric and nonnumeric). Appropriate templates are used to create various kinds of questions. A custom tagger is built, which takes the data entry and assigns it an entity, such that it can be generalized to a particular entity name. For uncategorized or unrecognized entities, options to add or modify entities are also provided. Preliminary experimental work carried out using this template system shows promising results. To the best of our knowledge, this is the first such question generation system based on structured data in the form of tables.

S. Totala

Papers

Automated Question Generation tool for structured data