Craft Data Science Insights: A 40-Minute Exploration – Session #2

Get ready to be inspired! This miniseries brings together five brilliant young minds from education, psychology, learning science, and survey methodology. They’ll share their innovative solutions to challenges powered by AI and data science.

Psychometrics Empowering Large Language Models in Chinese Essay Automated Scoring

Dr. Chanjin Zheng, East China Normal University

Psychometrics can empower large language models (LLMs) in assessment, both discriminative and ChatGPT alike. In this talk, I will share some research on the synergy of psychometrics and LLMs in Chinese essay automated scoring and evaluation. A developed Chinese essay automated scoring system will be introduced using the design-based research (DBR) method. I will further discuss the possibilities for combining psychometric devices and LLMs and demonstrate how psychometric devices can be used together with various LLMs to improve the design of an essay scoring system with several features, including essay scoring, text feature evaluation, feedback generation, and delivery. Then I’ll demonstrate how these three collaborate on more complex tasks, such as essay tutoring. Finally, I will introduce a new working framework, LCEAAG (Learning Copilot Enabled by Accurate Assessment and ChatGPT), together with ongoing research topics, such as interpretable essay automated scoring for higher-order essay writing skills.

View event recording: https://youtu.be/2AmBLN0C5m8?si=R-RS7KCa59HrY1O-

About the Presenter

Dr. Chanjin Zheng is an associate professor of educational psychology with a joint appointment at the Department of Educational Psychology of the Faculty of Education and the Shanghai Institute of Artificial Intelligence for Education (SH-iAIE) at the East China Normal University. In 2015, he completed a Ph.D. in Educational Psychology specializing in Educational Measurement and Statistics and a Master’s in Statistics at the University of Illinois at Urbana-Champaign in the United States. His primary research interests lie at the intersection of technology, assessment, and learning, particularly intelligent assessment and learning driven by artificial intelligence. He has directed or served as a technical advisor on several projects with Microsoft Research Asia, HanKao (HSK) International Education Technology (Beijing) Co., Ltd., Oxford University Press and iHuman Education.

Dr. Chanjin Zheng
Associate Professor
Department of Educational Psychology
Shanghai Institute of AI for Education
East China Normal University

Contact: Hua-Hua Chang, chang606@purdue.edu