A Comparative Study of ML and DL Approaches for Twitter Sentiment Classification

This research made use of various machine learning (ML) and deep learning (DL) methods — such as support vector machines, random forests, logistic regression, naive Bayes, and XGBoost, convolutional neural networks (CNNs), and feedforward neural networks (FNNs) — for tweet analysis to investigate public sentiment towards Ola and Uber. The objective is to determine the most effective method for distinguishing between good and negative tweets. Feature engineering techniques improve the algorithms interpretation of tweet content. To balance out the disparity between positive and negative tweets. The project aims to uncover customer wants and concerns on Twitter to help Ola and Uber, in addition to improving Algorithms Accuracy. The study intends to help these ride-hailing businesses make educated modifications to boost customer happiness by closely examining tweets. Essentially, the study assesses how well various ML and DL algorithms comprehend user feedback on Uber and Ola. The overarching goal is to not only enhance computational methods but also contribute to the improvement of these ride-hailing services, ultimately fostering a more positive online environment for Ola and Uber enthusiasts. In summary, the study investigates sentiment analysis techniques on Twitter to optimize understanding of Ola and Uber-related tweets, aiming to facilitate positive changes for the ride-hailing services and their customers, promoting a friendlier Twitter community.