Exploring Sentiment in Educational Tweets: Harnessing NLP and Wordnet for Deeper Insights

In order to illustrate how online educational services are offered, this article makes use of service blueprint technology. This diagram displays the online learning platform's internal service process, the customer contact point, and the service's user-facing components. Using this top-down approach when defining services helps to minimize difficulties connected with transience. We suggested combining NLP with WORDNET to do sentiment analysis on educational tweets. The dataset was normalized using Natural Language Processing (NLP) and WORDNET POS tagging. In addition, the Principal Component Analysis approach was used for Feature Reduction. During Ensemble Feature Selection, elastic net, recursive feature, and hybrid machine learning (decision tree (DT) and random forest (RF)) were used. Finally, as ML classification algorithms, Linear Support Vector Machine (SVM), Gaussian Naive Bayes (GNB), and Linear Regression (LR) were used. The experimental findings with various performance indicators have been shown. © 2024, Ismail Saritas. All rights reserved.

Author Keywords

Index Keywords