Abstract
Two types of students’ performance evaluation mechanisms are called formative and summative. In order to have productive results at the summative evaluation, the students must have had frequent constructive feedbacks throughout a course. It is not an easy task for a teacher to give frequent feedback. For that, online tests can be introduced with feedback. Constructive feedback should at least consist of Appreciation, Advice, and Evaluation. When incorporating the appreciation in feedback, the emotions can be analyzed to appreciate the effort of the student. The most feasible way to analyze emotions is through facial expressions. However, there should be a systematic way to track emotions based on each question. This research proposes a time interval based on the transient emotion peak considering the time taken to read a question as to the stimulus activity for an emotional change, to capture the emotional changes. A series of photographs were captured using a webcam for each question and analyzed the emotions using MS-emotion-API and a single photograph was selected based on the transient emotion peak, tracking the most different emotion. A video was recorded simultaneously and it was analyzed by a psychiatrist and kept as the benchmark dataset. Cross-validation test was done based on the output of API and the benchmark dataset. It was noticed that the sensitivity of the calculations if an image was detected within the defined time that emotion would be recognized correctly at a 93.52% accuracy. Specificity indicated that 70.83% chance of having a misclassification.
Presenters
Udeni Jayasinghestudent, university of colombo school of computing, Sri Lanka Ajantha Atukorale
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
KEYWORDS
Online education systems, Transient Emotion Peak, Online feedback, Emotions