Özet:

Abstract Learning is a lifelong process that allows individuals to acquire knowledge, skills, and attitudes through various experiences. In the recent pandemic times, there was a transformation in ways the society was acquiring knowledge and taking up education. There has been an exponential growth in number of e-learners attending classes in synchronous and asynchronous e-learning platform. The launch of e-university will benefit the e-learners to continue learning. This has created a need for evaluating the ecosystem of e-learning, the learning platform, learning analytics, e-learners satisfaction, quality of academic programs offered by the university and its reputation. The attention of e-learners based on reviews in terms of desired field of study, faculty expertise, research opportunities, and the curriculum structure.   The key attributes of e-learning are engagement, assessment, relevance, reflection, personalized learning recommendations and continual learning. This has opened an avenue for research to analyze the reviews of students to evaluate the e-learning platforms based on learning outcomes achieved. Most Challenging task is to find the perspective of the e-learners’ emotions from the huge data of the e-learners reviews. Text data gives qualitative information and this actionable knowledge can be quantified. The reviews on all e-learning platforms are mostly textual and this qualitative data needs to be quantified for analysis. There is a necessity to propose contextual emotion detection of e-learners by extracting the relevant information. Machine learning algorithms have revolutionized the text mining to get insights from diverse and huge dataset. This paper leverages machine learning techniques Multilayer Perceptron (MLP), Logistic Regression (LR), Random Forest (RF), Decision Tree (DT) used in emotion detection and analysis of e-learners to correlate the student satisfaction index are evaluated using E-Learners Academic Reviews (ELAR) dataset.  The DT and RF models consistently had high precision and accuracy scores of more than 90% in all academic emotion categories of excitement, happy, satisfied, not satisfied and frustration. This research also highlights the advantages of estimating the emotions of e-learners to evolve an e-learning platform that is conducive to their retention and satisfaction.

Anahtar Kelimeler: