Comprehension questions generation in Reading Progress assignments

What kind of questions can be created?

Educators can easily create comprehension questions in various formats, such as:

• Multiple Choice

• True/False

• and Open-ended questions

Based on the existing reading passage. Both the answers and distractors for these questions are automatically generated, though educators can edit them as needed.

How do you ensure content is safe and appropriate for students?

Content moderation is used to ensure that the generated content is safe and appropriate for the students' reading level. This feature is a valuable tool for online learning, distance education, and virtual classrooms, enhancing the capabilities of digital education. It supports blended learning, hybrid learning, and various e-learning platforms, providing interactive learning and online resources for educators.

By integrating AI into educational technology (EdTech), educators can streamline the creation of online homework, online assessments, and online tutoring materials. This feature is also beneficial for self-paced learning and adaptive learning, supporting both asynchronous learning and synchronous learning environments. Overall, Reading Progress enhances online student engagement and improves the effectiveness of remote learning.

How does the comprehension question generation technology work?  

Comprehension generation is powered by a machine learning model called GPT (Generative Pre-trained Transformer), designed to enhance online learning and digital education. Trained on vast amounts of text from the internet, the GPT model takes a reading passage and the number of requested questions as input to generate a variety of question formats and answers.

The questions formats that are generated include: 

• Multiple choice questions, including one correct answer and three distractors each

• Yes/No questions, including the correct answer

• Open ended questions, including short-form answer

The model intelligently selects which question formats to include based on the passage, ensuring a comprehensive evaluation. Not all formats or the entire passage may be covered in the generated questions, but the variety enhances interactive learning and student engagement in virtual classrooms and distance education.

Content moderation and quality safeguards are in place to ensure the questions are appropriate and relevant for students, making it a valuable tool for blended learning, hybrid learning, and various e-learning platforms. This feature supports remote learning by providing educators with the ability to create personalized reading assignments and online assessments, improving the effectiveness of self-paced learning, adaptive learning, and asynchronous learning environments. The integration of GPT in educational technology (EdTech) enhances online resources, making it a crucial component of modern digital literacy and online tutoring.

How should educators use the generative comprehension questions?  

Educators can leverage generative comprehension questions to quickly create comprehension evaluation opportunities for their students, enhancing online learning and digital education. This tool supports educators in virtual classrooms, distance education, and e-learning platforms by providing a quick and efficient way to generate multiple question formats for reading assignments.

It is essential for educators to review and edit the generated content for accuracy and appropriateness before assigning it to students. Although various filters and safeguards are employed to minimize questionable or inappropriate content, the AI technology, trained on a wide range of internet sources, requires educators to ensure the suitability of the questions for their students.

This feature is a valuable addition to blended learning, hybrid learning, and remote learning environments. It supports interactive learning, self-paced learning, and adaptive learning by providing personalized reading assignments and online assessments. The responsibility of content moderation falls on educators, making it crucial for maintaining the quality and relevance of online resources and educational technology (EdTech).

By confirming the appropriateness of the generated content, educators can enhance digital literacy, improve online student engagement, and ensure the effectiveness of online tutoring and online homework. This tool integrates seamlessly into various learning management systems (LMS) and virtual learning environments (VLE), supporting asynchronous and synchronous learning, online courses, and other e-learning initiatives.

What safeguards are employed while generating the questions and answers?  

• Educators are instructed to review and approve the comprehension questions and answers before sharing with their students.

• The comprehension questions and answers are on the passage provided by the educator.

• There is a limit of 10 questions that can be generated.

• Content moderation is employed on the passage, questions and answers before presented to educator.

• Generated questions and answers are only saved as part of the assignment and are not stored in other locations. 

Model limitations  

• Up to 10 questions can be generated per passage.

• Question types are selected by the model and cannot be guaranteed.

• The generation feature is designed to work on reading passages, and may not work as well for very long content (above 3000 words)

• In cases where high number of questions are requested for very short paragraphs, the model may return lower number of questions then requested

• Although dozens of languages are supported, translation errors may occur.

Data classification for the generated passage  

The text generated by the AI is owned by the educator using the functionality. For more information on the use limitations and data access and use terms, see the Microsoft Customer Agreement program under “Azure OpenAI Service."