Measuring Student Happiness Levels from Free-form Course Feedbacks

This paper presents an algorithmic design for opinion extractor from free-form course feedback texts. An unsupervised lexicon-based based algorithmic formulation is designed and implemented for identifying opinion polarity from free-form textual feedbacks written by students. The performance of the algorithm is evaluated on two different course feedback datasets obtained from ratemyprofessor website. The algorithm performance level observed is reasonably good. The algorithm produces easy to visualize graphical opinion summary from course feedbacks. This framework can be used in a course feedback system for automated interpretation of course feedbacks written by students.