Abstract
A key economic move that resulted in heated disputes was India's introduction of the Goods and Services Tax (GST). Social media channels offered a widely used forum for the people to express their views on the GST, providing insightful data for gauging mood and guiding next revisions. The emotion of 5629 GST-related tweets was assessed using the VADER lexicon after being obtained using the Twitter Developer API. The tf-idf feature was used for text vectorization, with 80% of the data going toward training and the remaining 20% going toward testing. In this study, six well-known classifiers—the Ridge Classifier, Logistic Regression, Linear SVC, Perceptron, Decision Tree, and K-Nearest Neighbor—were thoroughly compared to evaluate their performance in a range of circumstances. Accuracy, precision, recall, f-score, training, and testing times were all included in the performance measurements. The study presented novel pre-processing methods and examined the training/testing times before coming to the conclusion that the Ridge Classifier performed better than the others in terms of accuracy, precision, and efficiency. In this study, six well-known classifiers—the Ridge Classifier, Logistic Regression, Linear SVC, Perceptron, Decision Tree, and K-Nearest Neighbor—were thoroughly compared to evaluate their performance in a range of circumstances. Accuracy, precision, recall, f-score, training, and testing times were all included in the performance measurements. The study presented novel pre-processing methods and examined the training/testing times before coming to the conclusion that the Ridge Classifier performed better than the others in terms of accuracy, precision, and efficiency.

This publication has 13 references indexed in Scilit: