Towards a New Indicator for Evaluating Universities Based on Twitter Sentiment Analysis

Due to the rapid growth of social media platforms as a medium for providing opinions such as Twitter, a tremendous amount of informal statements based on opinions for various university related topics are created. Investigating tweet sentiments in the context of educational institutions can act as a vital complementary source to compare and evaluate universities. This paper deploys sentiment analysis methods to analyze the collected tweets about four chosen universities in Michigan, where each tweet conveys an opinion or feedback. The tweets were classified according to their sentiment into "Positive," "Negative," or "Neutral" using Machine Learning classifiers. Accordingly, the percentage of "Positive" tweets was used to carry out a comparison between the chosen universities. The results can be utilized by institutions policy makers to improve their educational environments. In addition, universities comparison and evaluation results can be enhanced using such vital indicators.