Institute of Information Science, Academia Sinica

Understanding how personality influences learning is key to optimizing programming education. In this talk, I present a series of studies exploring how physiological signals, note-taking behaviors, and system operation patterns can be leveraged to assess personality traits and adapt instructional strategies accordingly. Using wearable sensors, we found significant correlations between heart rate variability, galvanic skin response, and Big Five personality traits—especially extraversion and conscientiousness. Further, by tailoring programming materials based on these traits, students demonstrated significantly greater learning gains. In a parallel study, we analyzed students’ note-taking and programming behaviors in a front-end development course and achieved over 89% classification accuracy in predicting conscientiousness and openness using a two-layer random forest model. These results reveal a promising direction toward objective, scalable, and adaptive programming education based on students' individual learning styles and personality traits.

Chun-Hsiung received his B.S. in computer science from National National ChengChi University, and received both M.S. and Ph.D. in computer science from National Taiwan University. He was a research assistant of Institute of Information Science, Academia Sinica in 2003-2010. He was a faculty member of Department of Computer Information and Network Engineering, Lunghwa University of Science and Technology in 2010-2013. His current position is a faculty member of Department of Communications Engineering, Yuan Ze University. His research interests include big data analysis, crowd intelligence, e-learning systems, and Web information extraction.