Brandeis professor James Pustejovsky (COSI/LING) is currently working on a software titled the Diana program, which is an artificial intelligence (AI) program meant to aid teachers in seeing which students are engaged during class. According to the article, “TeachFX and Diana have a common goal of making sure the voices of all students are heard and that the message from teachers is getting through to everyone in the class.”
“We think this will be incredible assistance to the teacher, just to keep track of what is going on. To help the lesson proceed and the lesson plan you want,” said Pustejovsky, according to a Fox5 New York article. The purpose of the software is to enhance human connection, since students learn while interacting with others. The developers are “trying to close the opportunity gap between students of color, different ethnicities, even gender,” according to Jamie Plotkin, one of the developers of the program.
AI is helping students with how often they are speaking and how involved they are. “The goal is to point out the students who, for whatever reason, are not as involved as others,” according to the article.
Over the last few years, they have been developing “technology that can parse through the audio of the classroom,” said Plotkin. One of the features of the programs is called TeachFX, which was developed by Jamie Plotkin.
TeachFX is an app that can measure student engagement, which will allow teachers to see which of their students are following along and which students are falling behind. The methodology behind this is “providing data on how often a student interacts and what questions prompted that interaction,” according to the article.
A fuller version of the Diana software will soon be available for use: “What we focused on with Diana is to try to create an environment that is a 3D simulation with a human or another robot interacting with one another … it’s done by recreating [the student’s] environment through multiple cameras and she’s able to see [the student’s] face.”
According to Pustejovsky, the next step of the project will be engaging with students, so it can encourage those that are not participating to engage in the class.
The technology is still in its early stages of development, but numerous schools have already started implementing it in various programs.
Pustejovsky’s work is focused on modeling human communication with each other in computers and robots, according to his faculty page. In his work, Pustejovsky examines ways in which humans communicate through language, gestures, gaze, facial expression and action. According to his faculty page, Pustejovsky’s work involves developing computational simulations, theoretical and computational modeling of language and computational support for digital humanities.
Pustejovsky is a TJX Feldberg Professor of Computer Science at the university. He has taught Modal, Temporal and Spatial Logic for Language (COSI 112), Fundamentals of Natural Language Processing II (COSI 114), Introduction to Big Data Analysis (COSI 129), Computational Semantics (COSI 135), Information Extraction (COSI 137), Discourse and Dialog (COSI 233) and Topics in Linguistics (LING 190).
Pustejovsky has received numerous awards for his work including being a lecturer and keynote speaker at many institutions as well as being featured in multiple publications, according to his faculty page. He got his bachelor’s degree at the Massachusetts Institute of Technology, and his Ph.D. from the University of Massachusetts Amherst.