User reviews on social networking platforms like Twitter, Facebook, and Google+, etc. have been gaining growing interest on account of their wide usage in sentiment analysis which serves as the feedback to both public and private companies, as well as, governments. The analysis of such reviews not only plays a noteworthy role to improve the quality of such services and products but helps to devise marketing and financial strategies to increase the profit for companies and customer satisfaction. Although many analysis models have been proposed, yet, there is still room for improving the processing, classification, and analysis of user reviews which can assist managers to interpret customers feedback and elevate the quality of products. This study first evaluates the performance of a few machine learning models which are among the most widely used models and then presents a voting classifier Gradient Boosted Support Vector Machine (GBSVM) which is constituted of gradient boosting and support vector machines. The proposed model has been evaluated on two different datasets with term frequency and three variants of term frequency-inverse document frequency including uni-, bi-, and tri-gram as features. The performance is compared with other state-of-the-art techniques which prove that GBSVM outperforms these models.

Deepfake text known as synthetic text, involves using artificial intelligence (AI)-generated text to create fabricated information or imitate actual individuals. Twitter tweets related to deepfake can be used for many malicious intents, including impersonation, creating fake news, and spreading misinformation. The main goal of this investigation is to detect people’s sentiments related to deepfake technology with an advanced technique. A novel sentiment majority voting classifier (SMVC) is proposed for the sentiment labeling of collected tweets. The proposed SMVC selects the final sentiment from three lexicon-based models TextBlob, valence-aware dictionary and sentiment reasoner (VADER), and AFINN using a majority voting mechanism. For sentiment classification, we propose a novel transfer feature where embedding features are fed to a long short-term memory (LSTM), and decision tree (DT) models and outputs are combined into a single feature set. Extensive experiments show that transfer learning-based feature engineering results in the highest performance. The logistic regression outperforms with the highest accuracy of 98.9% with minimum computational complexity. The sentiment classification performance of each applied model is validated using the k-fold cross-validations. Moreover, performance assessment with existing state-of-the-art models is also carried out to show the robustness of the proposed technique.