Enhancing Online Education Experience Using Learners' Comments: A Novel Approach To Feedback Analysis

Abstract

With the rise of e-learning platforms, approaches to education have changed and never before allowed for such flexible and accessible learning options. Optimizing these platforms' effectiveness and personalisation is still difficult, though. The incorporation of classification as a revolutionary method for enhancing e-learning experiences is examined in this research. The study intends to identify and analyze learners' emotional states, views, and attitudes inside digital information and interactions by utilizing a new approach for sentiment analysis. The study's technique includes applying classification logistic regression, random forest classifier and SVM.

Through normalization to learner-generated content in an online learning environment. This approach offers insights into how to modify instructional tactics to meet the emotional requirements of students by examining correlations between sentiments expressed and learning results. The results show that using SMOTE for data balance enhances classifier performance and provides a reliable way to glean useful information from learner comments. This methodology not only augments comprehension of student attitudes but also furnishes a basis for enhancing the caliber of virtual learning via well-informed decision-making. To improve the analysis's resilience, more research may entail adjusting the model's parameters and looking into new features. This research article investigates a comparative study of logistic regression, random forest classifier and SVM are done