MS in Computer Engineering (M.S.C.P.)
Degree Granting Department
Computer Science and Engineering
Tempestt Neal, Ph.D.
Shaun Canavan, Ph.D.
Lawrence Hall, Ph.D.
Affective computing, In-group advantage, Physiological data, supervised learning, unsupervised learning
Research shows that emotional distress has a statistically significant impact on a student’s grade point average and intent to drop out of college. Because students of different races have varying college experiences, it is important to understand the emotional experiences of different racial groups to better support students’ needs and academic success. In this work, we explore several physiological responses to ten different emotional stimuli captured from 140 students. We employ unsupervised learning via the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm and supervised learning via Random Forests and Support Vector machines to analyze clustering partitions and classification performance according to emotional state (e.g., happiness), race (e.g., Asian), and all combinations of the two (e.g., happiness and Asian). We also consider a much broader scope by analyzing clustering partitions and classification results according to subject ID (i.e., biometric identification) and gender.
Significant findings show that blood pressure and respiration rate provided more accurate partitions of the data with respect to emotion and race, in addition to gender. The emotional states of sadness and startled were classified with 99% and 94%, respectively, while surprise and skepticism were often misclassified as the other (i.e., surprise as skepticism and vice versa). However, in general, both unsupervised and supervised learning tasks suggested significant overlap in physiological emotional responses across the different racial groups.
Scholar Commons Citation
Zanna, Khadija, "Toward Culturally Relevant Emotion Detection Using Physiological Signals" (2020). USF Tampa Graduate Theses and Dissertations.