Helping Teachers Create Course Materials

 

It may surprise you to hear that teachers spend 20-30% of their time creating and modifying course materials. For example, a teacher might know that their students are interested in pollution from a local chemical plant, and so find news stories to bring into a lesson on cellular biology. However, teachers face a big problem using non-textbook materials like the example above: they have to create test questions for assessments, and they have to create instructional materials like worksheets and concept maps.

Our recent project in the Learner Data Institute at the University of Memphis is trying to solve these problems by giving teachers the tools they need to create custom course materials. Our approach starts with the text that teachers want to use, which could be a news article, a short story, or even a book. We use a technology called deep neural networks to extract knowledge from the text, then turn that knowledge into test questions and instructional materials.
For example, if the teacher wanted to create a fill- in- the- blank question from a text on cellular biology, our system would select sentences that are the most important in the overall understanding of the text and then create a fill in the blank question like this:

The _________ at the distal end of the axon is rich in mitochondria and contains many tiny vesicles that store neurotransmitter.

The teacher can then think of incorrect student answers to the question and feed those into the system to get explanations for why the answer is incorrect. For example, if the teacher gave the incorrect answer “acetylcholine”, the system would respond with:

Acetylcholine is not right. The correct answer is cytoplasm. Acetylcholine is synthesized in the cytoplasm of nerve terminals by the enzyme choline acetyltransferase, and is then transported into synaptic vesicles.

Our system can also generate WH-questions like “What stores neurotransmitter”, and create multiple choice responses with plausible distractor items.
This work is still in an early stage, but we’d love to work with teachers and get feedback. If you are a teacher and would like collaborate, please contact me at aolney@memphis.edu

Andrew McGregor Olney

Andrew M. Olney presently serves as Professor in both the Institute for Intelligent Systems and Department of Psychology at the University of Memphis. Dr. Olney received a B.A. in Linguistics with Cognitive Science from University College London in 1998, an M.S. in Evolutionary and Adaptive Systems from the University of Sussex in 2001, and a Ph.D. in Computer Science from the University of Memphis in 2006. His primary research interests are in natural language interfaces. Specific interests include vector space models, dialogue systems, unsupervised grammar induction, robotics, and intelligent tutoring systems.

Leave a Reply

Your email address will not be published. Required fields are marked *