AI chatbot to increase cultural relevancy of STEM lessons, engage marginalized students
Oct 17, 2023
Indiana University researchers are developing a new chatbot that will make it easier for teachers to create more inclusive STEM activities — opening the door for marginalized students to engage in sustained STEM learning and STEM careers.
The project, led by IU’s Jeremy Price and Sunandan Chakraborty, will leverage artificial intelligence to engage teachers in conversations about the best ways to adapt existing lesson plans and create new ones that are more inclusive and equitable for diverse students. Their work is supported by a one-year, $200,000 from the National Science Foundation.
“STEM lesson plans in particular can be very difficult for teachers to make more culturally relevant for their students,” said Price, an assistant professor of technology, innovation and pedagogy in urban education at the School of Education at IUPUI and project director of the Collaborative for Equitable and Inclusive STEM Learning. “This chatbot will bring equity, inclusion and rigorous STEM to the forefront, providing important suggestions, pointers and resources to teachers to better engage with students, build a sense of community and prepare them for a life of STEM learning.”
While the initial work is focused on K-12 teachers, the researchers hope students ultimately benefit the most. Introducing students to STEM learning activities that are relevant to them may get them more excited about future STEM learning and the various pathways it can lead them on in their lives and future careers. The researchers said this is particularly important for students of diverse backgrounds, who are often underrepresented in postsecondary science and engineering degrees.
The researchers, including research assistants from Luddy’s M.S. in Applied Data Science program, will ensure the chatbot’s cultural sensitivity through community engagement. Building on the Collaborative for Equitable and Inclusive STEM Learning’s existing outreach and relationships in Indianapolis’ Near Eastside community, the chatbot will be trained using data from teachers and community members in the neighborhood who understand the importance of equity and inclusion in schools.
The chatbot’s ability to offer a customized experience also makes it unique.
“One of the chatbot’s most distinctive features will be its capacity to adapt and learn from the diverse backgrounds and preferences of users, tailoring interactions and activities with a high degree of context-sensitivity,” said Chakraborty, an assistant professor of data science in the Luddy School of Informatics, Computing and Engineering at IUPUI. “This adaptive and personalized approach sets it apart, ensuring that it can effectively cater to the unique needs and characteristics of each participant.”
The researchers will emphasize universal design to make the chatbot accessible to everyone, including those with disabilities. Additionally, they will build upon Chakraborty’s expertise with developing AI models and novel Natural Language Processing that can be used in real-world scenarios to develop more sophisticated, user-interaction-focused language models.
To collect the necessary data to train the chatbot, affiliated faculty, staff and graduate students from the Collaborative for Equitable and Inclusive STEM Learning are beginning their work with the community this month. Focus groups will be convened to understand the ideas and aspirations that families and educators have for their students in STEM, with that input being used to train the chatbot.
An early beta version of the chatbot will be ready for teachers to use within the next 12 months.